Abstracts

Transcription

Abstracts

Pre-Conference Tutorials — Room 2000
8:00 a.m.
8:00 a.m.
Optimizing Big Data Programs
Model Selection with SAS/STAT® Software
This seminar will review traditional SAS techniques for optimizing programs
that require table lookups. We will do this by focusing on in-memory
techniques. The techniques discussed will include: Arrays (one-dimensional
and multi-dimensional). Hash objects. Formats. We will look at indexing as a
means of “super-charging” your lookups and sampling for the purpose of
exploring your data. We will also investigate several newer features in SAS
as a way to optimize your programs in a BI environment, including grid
computing and the use of PROC SCAPROC to help “gridify” your programs.
This is an extra fee event. See registration.
When you are faced with a predictive modeling problem that has many
possible predictor effects – dozens, hundreds or even thousands – a natural
question is, "What subset of the effects provides the best model for the
data?" This tutorial explores how you can use model selection methods in
SAS/STAT software to address this question. Although the model selection
question seems reasonable, trying to answer it for real data can create
problematic pitfalls, leading some experts to stridently denounce model
selection. This workshop focuses on the GLMSELECT procedure and shows
how it can be used to address and mitigate the intrinsic difficulties of model
selection. You will learn how to:
Christine Riddiough, SAS
8:00 a.m.
Modeling Categorical Response Data
Maura Stokes, SAS
Logistic regression, generally used to model dichotomous response data, is
one of the basic tools for a statistician. But what do you do when maximum
likelihood estimation fails or your sample sizes are questionable? And
wWhat happens when you have more than two response levels? And how
do you handle counts? This tutorial briefly reviews logistic regression for
dichotomous responses, and then illustrates alternative strategies for the
dichotomous case and additional strategies such as the proportional odds
model, generalized logit model, conditional logistic regression, and Poisson
regression. . The presentation is based on the third edition of the book
Categorical Data Analysis Using the SAS System by Stokes, Davis and Koch
(2012). An existing working knowledge of logistic regression is required for
this tutorial to be fully beneficial. This is an extra fee event. See registration.
8:00 a.m.
A-to-Z Analysis and Validation Using PROC Tabulate
Sunil Gupta, Gupta Programming
Generate most any combination of summary table layout! Do you know up
to seven different table layouts, five ways to control order, three ways to
include missing data, or four ways to include subtotals? After this course,
you can expect to take advantage of PROC Tabulate’s powerful feature to
group and analyze data in most any summary table layout. This unique
course explores core syntax options for up to 26 key summary table
structures, including mixing both continuous and categorical data with
ODS. By applying the simple SAS examples provided throughout the
course, you too can master PROC Tabulate in your daily programming
environment. Each student receives the companion SAS e-guide, which is a
great reference tool for searching, cutting and pasting concise model SAS
examples. This is an extra fee event. See registration
Funda Gunes, SAS
• Use extensive model selection diagnostics including graphics to detect
problems.
• Use validation data to detect and prevent both under- and over-fitting.
• Use modern-penalty based methods, including LASSO and adaptive
LASSO, as alternatives to traditional methods such as stepwise selection.
• Use bootstrap-based model averaging to reduce selection bias and
improve predictive performance. This tutorial requires an understanding
of basic regression techniques. This is an extra fee event. See registration
8:00 a.m.
How to Become a Top SAS Programmer
Michael Raithel, Westat
This groundbreaking seminar, based on the new SAS Press book with the
same title, provides clear-cut strategies for becoming a top SAS
programmer. Whether you are a student or a statistician, a programmer or a
business analyst, this seminar shows how you can streamline and revitalize
your career to become the top SAS professional in your organization.
Instructor Michael A. Raithel will reveal how to unleash the SAS expert
within you by learning how to take advantage of a wide variety of proven
job strategies that use SAS to help you to maximize your knowledge, skills,
accomplishments, recognition and pay. Featuring lectures, discussions and
paper-and-pen exercises, the seminar details key SAS programming
fundamentals to master, strategies for becoming the go-to SAS person in
your organization, ways to become involved and recognized in the greater
SAS community, and how to exploit the best sources of SAS information.
You will leave the seminar with a written set of objectives and the
knowledge to implement them. Armed with the information from this
course, your set of goals and objectives, and your own ambition, there are
no limits to how far you can go with your SAS programming career. This is
an extra fee event. See registration.
www.sasglobalforum.org/2013
1
8:00 a.m.
10:30 a.m.
SAS Enterprise BI – Tables, Maps, and Cubes:
Understanding the Differences
Introduction to the MCMC Procedure in SAS/STAT
Software
Eric Rossland, SAS
Fang Chen, SAS
You have several choices of data sources when analyzing data and creating
reports with the SAS BI applications. Whether you’re new to SAS BI or have
been using it since the beginning, this seminar will help you to understand
the different types of data sources. Tables, OLAP cubes, and information
maps can all provide data to the SAS Add-In for Microsoft Office, SAS
Enterprise Guide, and SAS Web Report Studio. Each of these data sources
has unique characteristics that can enhance the analysis and reporting. Join
us as we investigate each data source and how it can be used in the various
SAS applications. Intended Audience: Anyone who has SAS Enterprise BI
Server or wants to learn more about the different data sources as well as
the analytic and reporting capabilities they provide. This is an extra fee
event. See registration.
The MCMC procedure is a general-purpose Markov chain Monte Carlo
simulation tool designed to fit Bayesian models. It uses a variety of
sampling algorithms to generate samples from the targeted posterior
distributions. This workshop will review the methods available with PROC
MCMC and demonstrate its use with a series of real-world applications.
Examples will include fitting a variety of parametric models: generalized
linear models, linear and nonlinear models, hierarchical models, zeroinflated Poisson models, and missing data problems. Additional Bayesian
topics such as sensitivity analysis, inference of functions of parameters, and
power priors will be discussed, and applications will be demonstrated with
the MCMC procedure. This workshop is intended for statisticians who are
interested in Bayesian computation. Attendees should have a basic
understanding of Bayesian methods and experience using the SAS
language. This tutorial is based on SAS/STAT 12.1. This is an extra fee event.
See registration
10:30 a.m.
Creating Statistical Graphics in SAS
Warren Kuhfeld, SAS
Effective graphics are indispensable in modern statistical analysis. SAS
provides statistical graphics through ODS Graphics, functionality that is
used by statistical procedures to create statistical graphics as automatically
as they create tables. ODS Graphics is also used by a family of Base SAS
procedures designed for graphical exploration of data. This tutorial is
intended for statistical users and covers the use of ODS Graphics from start
to finish in statistical analysis. You will learn how to:
• Request graphs created by statistical procedures.
• Use the SGPLOT, SGPANEL, SGSCATTER and SGRENDER procedures in
SAS/GRAPH® to create customized graphs.
• Access and manage your graphs for inclusion in Web pages, papers and
presentations.
• Modify graph styles (colors, fonts and general appearance).
• Make immediate changes to your graphs using a point-and-click editor.
12:30 p.m.
Beyond the Basics: Advanced Macro Tips and
Techniques
Jim Simon, SAS
This seminar will investigate ways to automate common programming
tasks through the use of advanced macro techniques. We will investigate
different techniques for generating repetitive data-driven macro calls,
including the use of the EXECUTE routine. We will look at how to use
various I/O functions to access SAS data sets and create your own macro
functions. We will also examine expediting the process of importing
external files such as CSV and Excel files. Along the way, we will learn
techniques for validating user input. This is an extra fee event. See
registration.
• Make permanent changes to your graphs with template changes.
• Specify other options related to ODS Graphics. This is an extra fee event.
See registration
12:30 p.m.
What Will DS2 Do for You?
Mark Jordan, SAS
DS2 is an exciting, powerful new programming language available in SAS
9.4. It enables users to explicitly control threading to leverage multiple
processors when performing data manipulation and data modeling. It also
improves extensibility and data abstraction in your code through the
implementation of packages and methods. Paired with the SAS Embedded
Process, DS2 enables you to perform processing similar to SAS in
completely new places, such as in-database processing in relational
databases, the SAS High-Performance Analytics grid and the DataFlux®
Federation Server. In this seminar, you will learn the basics of writing and
2
executing DS2 code. Discover how you can capitalize on the power of the
new DS2 programming language in your own work. This is an extra fee
event. See registration.
columns, some of which can cause unexpected results for the unwary user.
And after all, PROC SQL is part of Base SAS; so, though you may need to
learn a few new keywords to become an SQL wizard, no special license is
required! This is an extra fee event. See registration.
12:30 p.m.
Output Delivery System: The Basics and Beyond
Kirk Paul Lafler, Software Intelligence Corporation
This course explores the various techniques associated with output
formatting and delivery using the Output Delivery System (ODS). Numerous
examples will be presented to command mastery of ODS capabilities while
providing a better understanding of ODS statements and options to deliver
output any way that is needed. Topics include:
• SAS-supplied formatting statements and options.
• Selecting output objects with Selection or Exclusion Lists.
• Formatting Output as RTF, PDF, Microsoft Excel, and HTML.
• Using the Escape character to enhance output formats.
• Exploring ODS statements and options.
• Implementing scrollable tables in HTML output with static column
headers.
• Enabling/disabling borders.
• Generating HTML hyperlinks in RTF output.
• Adding images to RTF output.
• Removing gridlines and shading in RTF output.
• Creating a printable table of contents in PDF output.
• Sending output to Microsoft Office.
• Constructing drill-down applications with the DATA step, ODS and SAS/
GRAPH software.
• Creating thumbnail charts.
• Techniques for creating user-defined ODS styles.
• An introduction to the customization of output with the TEMPLATE
Procedure. This is an extra fee event. See registration
12:30 p.m.
Demystifying PROC SQL
Christianna Williams, Independent Consultant
Subqueries, inline views, outer joins, Cartesian products, HAVING
expressions, Set operators, INTO clauses – even the terminology of SQL can
be rather daunting for SAS programmers who use DATA step for data
manipulation. Not to mention the profusion of commas and complete
dearth of semicolons found in a PROC SQL step! Nonetheless, even the
most die-hard DATA step programmers must grudgingly acknowledge that
there are some tasks – such as the many-to-many merge or the “not-quiteequi-join” – that would require Herculean effort to accomplish with DATA
steps. However, these tasks can be achieved amazingly concisely, even
elegantly, using PROC SQL. This seminar will present a series of increasingly
complex examples to illustrate the function of each of PROC SQL’s clauses,
with particular focus on summarization/aggregation and a variety of joins.
Additionally, the examples will illuminate how SQL “thinks” about rows and
3
4
Applications Development — Room 2014
4:00 p.m.
2:00 p.m.
Fueling the Future of an Energy Company
Stijn Vanderlooy, EDF Luminus
Floating on Cloud 9.3: Leveraging the Cloud with SAS®
and Google Drive
William Roehl, MarketShare
Paper 001-2013
Our organization has been utilizing Google Drive (previously Google Docs)
to keep project documentation centrally located for ease of access by any
user on any platform. Up to this point, SAS® developers had to manually
import or export data sets to or from flat files or Microsoft Excel in order to
update data stored in the cloud. This paper provides a powerful macro
toolkit that facilitates direct access to Google Spreadsheets through its
published API, allowing uploading, downloading, deletion, and live data
manipulation of cloud-based data. This provides an opportunity for a
significant reduction in the amount of manual work and time required for
SAS developers to perform these basic functions. Code was developed with
SAS 9.3 and HTMLTidy (25MAR2009) running under Microsoft Windows 7.
2:30 p.m.
Using PROC FCMP in SAS® System Development: Real
Examples
Xiyun (Cheryl) Wang, Statistics Canada
Yves Deguire, Statistics Canada
Paper 505-2013
This paper discusses the use of the FCMP procedure, focusing mainly on
two aspects. One aspect is that a lot of complex mathematical functions to
be used in our systems are difficult to implement using SAS® macros. I have
encapsulated these complex functions into PROC FCMP functions and used
them seamlessly in PROC OPTMODEL. Another aspect is the need to apply
generic message handling across different SAS components, such as the
DATA step, SAS macros, and PROC steps. PROC FCMP again becomes a
natural fit for this purpose that can be easily invoked by any SAS program
blocks.
3:00 p.m.
Create Your Own Client Apps Using SAS® Integration
Technologies
Chris Hemedinger, SAS
Paper 003-2013
SAS® Integration Technologies allows any custom client application to
interact with SAS services. SAS® Enterprise Guide® and SAS® Add-In for
Microsoft Office are noteworthy examples of what can be done, but your
own applications don't have to be that ambitious. This paper explains how
to use SAS Integration Technologies components to accomplish focused
tasks, such as run a SAS® program on a remote server, read a SAS data set,
run a stored process, and transfer files between the client machine and the
SAS server. Working examples in Microsoft .NET (including C# and Visual
Basic .NET) as well as Windows PowerShell are also provided.
Paper 004-2013
EDF Luminus is a producer and supplier of energy in Belgium. The company
is active in several markets for trading commodities (including electricity,
natural gas, and oil-related products). The market data modeling team is
responsible for all information related to these markets. Besides a daily
collection and verification of all the published prices, the team is
responsible for a diverse set of transformations and manipulations of these
data. Examples are numerous and include volatility estimation and price
forecasts. In this paper we present a successful application of a SAS® tool
developed in-house that is used by the market data modeling team to
support its core tasks. The tool runs autonomously three times a day.
4:30 p.m.
Macro Quoting to the Rescue: Passing Special Characters
Art Carpenter, CA Occidental Consultants
Mary Rosenbloom, Edwards Lifesciences, LLC
Paper 005-2013
We know that we should always try to avoid storing special characters in
macro variables. We know that there are just too many ways that special
characters can cause problems when the macro variable is resolved.
Sometimes, however, we just do not have a choice. Sometimes the
characters must be stored in the macro variable whether we like it or not.
And when they appear we need to know how to deal with them. We need
to know which macro quoting functions will solve the problem, and even
more importantly why we need to use them. This paper takes a quick look
at the problems associated with the resolution and use of macro variables
that contain special characters such as commas, quotes, ampersands, and
parentheses.
5:00 p.m.
Line-Sampling Macro for Multistage Sampling
Charley Jiang, University of Michigan
James Lepkowski, University of Michigan
Richard Valliant, University of Michigan
James Wagner, University of Michigan
Paper 007-2013
In the SAS® world, one tool, PROC SURVEYSELECT, is widely used for
probability sample selection. However, the procedures implemented in
PROC SURVEYSELECT are basic selection tools that must be assembled into
larger systems for complex probability samples, particularly multistage
samples. This paper describes the development and operation of a set of
sampling macros built around PROC SURVEYSELECT for sampling the
ultimate stage units in a multistage sample. Examples of several situations
where the macro can be most beneficial are also given.
5:30 p.m.
SAS® Analytics Optimized with Intel Technologies
Mark Pallone, Intel Corporation
(Invited) Paper 539-2013
Several server platform configurations and technologies have a direct
impact on performance and scalability that are critical for SAS® workloads.
Intel Cache Acceleration Solution (iCAS) reduces storage latency and
transparently accelerates Applications, Servers, and Virtual Machines. iCAS
5
was used to measure IO performance improvements for SAS mixed
analytics workload. Resulting IO performance metrics will be shared.
Performance testing was conducted on 3rd Generation Intel Core processor
family server platform that is expected to ship in Q3 2013. Comparison of
performance metrics between 2nd and 3rd Generation Intel Core processor
families based on mixed analytics workload will be reviewed. In addition,
some preliminary performance metrics on 4th generation Intel iCore
processor will be shared as well.
Beyond the Basics — Room 2016
10:30 a.m.
Using the SAS® Data Step to Generate HTML or TextBased Mark-Up
Matt Karafa, The Cleveland Clinic
The author presents macros which produce reports direct to MS Wordcompliant HTML, thus demonstrating an alternative method to create MS
Word documents from SAS®. The first step is to create a mock-up of the
table in an external mark-up editor, then use SAS to produce the text that
creates the file, interspersing the required data between the mark-up tags.
These macros demonstrate a way to increase the control and flexibility over
what is available via the traditional ODS RTF or HTML mechanism. Further,
via this method, any text-based mark-up language (HTML, RTF, LaTeX, etc.)
can be produced with a minimal effort.
11:30 a.m.
The Hash-of-Hashes as a "Russian Doll" Structure: An
Example with XML Creation
Joseph Hinson, MERCK
Paper 021-2013
SAS®9 hash objects have inspired novel programming techniques. The
recent discovery that hash tables can contain even other hash objects:
“hash of hashes” opens the door to their application to hierarchical data
processing. Because hierarchies, like Russian dolls, can be considered
"containers within containers.” Thus, nested hash objects could model XML,
a hierarchical data structure increasingly finding its way into clinical trial
data. Clinical programmers now have to deal with hierarchical as well as
tabular and relational data sets. SAS® now provides tools like the XML
libname engine and XML mapper. This paper aims to show, using a
simplified CDISC LAB model, that the hash object could well be another
tool for creating XML.
12:00 p.m.
Optimize Your Delete
Brad Richardson, SAS
Paper 022-2013
Have you deleted a data set or two from a library that contains thousands
of members, using PROC DATASETS? If so, you probably have witnessed
some wait time. To maximize performance, we have reinstated PROC
DELETE as a supported procedure. One of the main differences between
PROC DELETE and PROC DATASETS DELETE is that PROC DELETE does not
need an in-memory directory to delete a data set. So what does this mean
exactly? This presentation explains all. For PROC DATASETS DELETE fans,
there are optimizations, as well. Come to this presentation to learn more
from the developer.
6
2:00 p.m.
Using Mail Functionality in SAS®
Erik Tilanus, Synchrona
The DATA step can write information to virtually any destination. You are
probably familiar with writing to SAS data sets and external files. But also
email can function as a destination. The paper will discuss how to configure
the email features in the system options and share practical hints how to
use them. Then we will proceed with sending a simple email from the DATA
step, with or without attachments. Then we will use extensions to the
standard PUT statement to support the email facility to send personalized
mass-mailings. Finally, we will show how to send procedure output that has
been created using ODS.
3:00 p.m.
The Ins and Outs of Web-Based Data with SAS®
Bill McNeill, SAS
Paper 024-2013
Do you have data on the Web that you want to integrate with SAS®? This
paper explains how you can obtain Web data, process it and export it back
out to the Web. Examples will use existing features, such as the SOAP and
XSL procedures, the XML mapper application, and XMLV2 LIBNAME engine,
along with two new features: the XMLV2 LIBNAME engine AUTOMAP
option and the JSON procedure. The AUTOMAP option allows for creation
of default XML Mapper files within SAS. The JSON procedure exports SAS
data sets in JSON format to an external file. And if you need to write freeform JSON output, forget the SAS PUT statements; the JSON procedure
supports free-form JSON output as well.
4:00 p.m.
Internationalization 101: Give Some International Flavor
to Your SAS® Applications
Mickael Bouedo, SAS
Steve Beatrous, SAS
Paper 025-2013
Do you have SAS® users worldwide? Do you want your SAS application to
be useable in many languages? SAS® 9.4 internationalization features will
get you there efficiently. If you want to adapt your SAS application for
different cultures, SAS internationalization is the step which generalizes
your product to be language-independent. Internationalization features in
SAS include the ENCODING, LOCALE, and TIMEZONE options; the SASMSG
function; the NL formats; and many more features that help you to write
code once so it can run in different cultural environments with low
maintenance. This paper describes how to successfully internationalize
your SAS programs and make them ready for the world.
4:30 p.m.
ISO 101: A SAS® Guide to International Dating
Peter Eberhardt, Fernwood Consulting Group Inc
Xiaojin Qin, Covance Pharmaceutical Research and
Development CO., Ltd.
For most new SAS® programmers, SAS dates can be confusing. Once some
of this confusion is cleared, the programmer might then come across the
ISO date formats in SAS, and another level of confusion sets in. This paper
reviews SAS date, SAS datetime, and SAS time variables and some of the
ways they can be managed. It then turns to the SAS ISO date formats and
shows how to make your dates international.
5:30 p.m.
Census Retires PROC COMPUTAB
Christopher Boniface, U.S. Census Bureau
Nora Szeto, U.S. Census Bureau
Hung Pham, U.S. Census Bureau
Paper 027-2013
PROC COMPUTAB is used to generate tabular reports in a spreadsheet-like
format. PROC COMPUTAB has been around a long time. It has served us
well at Census, but it is time to replace it with reporting procedures that are
more modern. This paper shows you how to create hundreds of Excel tables
using ODS TAGSETS EXCELXP. We discuss how we converted PROC
COMPUTAB to both PROC TABULATE and PROC REPORT to create complex
Census tables. Moreover, how we use PROC TABULATE as the computing
engine to handle overlapping format ranges and PROC REPORT as the
reporting tool to create polished Excel tables. We reveal how to control the
appearance of the Excel tables including column widths, row heights, and
formats.
allocation decisions become significant. This paper describes how we are
using SAS® Enterprise Miner™ to develop a model to score university
students based on their probability of enrollment and retention early in the
enrollment funnel so that staff and administrators can work to recruit
students that not only have an average or better chance of enrolling but
also succeeding once they enroll. Incorporating these results into SAS® EBI
will allow us to deliver easy-to-understand results to university personnel.
12:00 p.m.
Using SAS® BI for Integrated Bank Reporting
James Beaver, Farm Bureau Bank
Paper 045-2013
This paper shows how Base SAS®, SAS® Enterprise Guide®, SAS/ETS®, and
SAS® BI are used to provide a comprehensive view of bank performance.
Data is extracted from the G/L, loan, deposit, and application systems, realtime data is accessed to provide up-to-the-minute results on loan activity,
and system reports are read in to provide additional information. PROC
COMPUTAB is used to create financial statements, OLAP cubes are used to
provide reports on bank balance sheet components and budget
comparisons on non-interest income and expense items by department,
and dashboards are used to provide real-time reports on loan originations.
The reports are presented using SAS BI through the SAS data portal to
provide real-time, trend, and historical reports on the bank’s performance.
Business Intelligence Applications — Room 2009
2:00 p.m.
10:30 a.m.
SAS® High-Performance Analytics: Big Data Brought to
Life on the EMC Greenplum Data Computing Appliance
SAS® Business Intelligence Development Roundtable:
SAS Business Intelligence Solutions Portfolio and Future
Focus
Greg Hodges, SAS
Stuart Nisbet, SAS
Don Chapman, SAS
James Holman, SAS
Oita Coleman, SAS
Tammi Kay George, SAS
Paper 063-2013
Join SAS executives from Product Management and Business Intelligence
Research & Development for an interactive panel discussion on the current
BI and reporting solutions portfolio including SAS Enterprise BI Server, SAS
Enterprise Guide, Mobile BI and SAS Visual Analytics. These experts will
answer questions such as how to decide which product to use when, offer
deployment and implementation best practices and also provide guidance
on strategies to add Visual Analytics to your mix of SAS solutions. High level
roadmaps will be shared by product management that cover the product
portfolio and there will be dedicated time for Q&A to make sure your
questions get answered!
11:30 a.m.
A Data-Driven Analytic Strategy for Increasing Yield and
Retention at Western Kentucky University Using SAS
Enterprise BI and SAS® Enterprise Miner™
Matt Bogard, Western Kentucky University
Paper 044-2013
Paul Cegielski, Greenplum
This presentation will describe the proof-of-concept project to apply highperformance analytics (HPA) to call center and other data in an effort to
quickly identify and act on customer service opportunities. Discussion will
include functionality and performance metrics of SAS® High-Performance
Analytics procedures, the new SAS® DS2 language, the fast-loading
capability of the Greenplum DCA, and the ability to deploy models built on
the DCA to other databases. Since some of the most valuable data is
unstructured, such as the free-form text notes entered by call center staff,
the presentation will describe how SAS® Text Miner is used in conjunction
with the HPA DCA to include unstructured data in analyses and modeling.
3:00 p.m.
SAS® Stored Processes Are Goin’ Mobile!: Creating and
Delivering Mobile-Enabled Versions of Stored Process
Reports
Michael Drutar, SAS
Paper 046-2013
SAS® BI programmers have been clamoring for a way to quickly create
mobile-enabled versions of existing SAS BI content. This can be difficult
because there may be multiple BI reports that need to be converted.
Fortunately, PROC STP (new in SAS 9.3) is a solution that can take a system’s
existing SAS Stored Process reports and create mobile-enabled versions of
them. This paper shows how PROC STP can capture the ODS HTML output
from a stored process, create an HTML file from it and email the file to any
mobile device (iPhone®, Android, etc.). Using this method, any existing SAS
Stored Process report’s output can be easily mobile-enabled.
As many universities face the constraints of declining enrollment
demographics, pressure from state governments for increased student
success, as well as declining revenues, the costs of utilizing anecdotal
evidence and intuition based on “gut” feelings to make time and resource
7
3:30 p.m.
Linking Strategy Data in BI Applications
these features to enhance basic OLAP cubes by using member properties,
defining dynamic measures and dimensions using the MDX language, and
improving performance for high data volumes.
Paper 050-2013
Data Management — Room 2001
Bharat Trivedi, SAS
Christiana Lycan, SAS
SAS® Enterprise BI is a key part of every strategy management
implementation, but it is not always easy to link strategy data with the
correct views of information in SAS Enterprise BI. And even when the
linking is straightforward, users may not be authorized to see all of the
information. As a result, several views of the same reports must be created
for linking. Users need an easier way to link and display the correct view of
strategy management content in SAS Enterprise BI. This paper reveals how
to link information created in SAS Enterprise BI that is context-aware and
secure.
4:00 p.m.
Self-Service Data Management: Visual Data Builder
Malcolm Alexander, SAS
Sam Atassi, SAS
Paper 051-2013
Successful data preparation is the key to extracting meaningful knowledge
from data. SAS® Visual Data Builder allows you to access data from
enterprise sources and transform it for use in business intelligence, data
visualization and data mining tasks. This paper discusses self-service data
management techniques available using SAS Visual Data Builder, as well as
the unique features enabling it to load data into SAS® LASR(TM)
Analytic(TM) Server.
4:30 p.m.
Stop your 'Wine”-ing: Use a Stored Process!
Tricia Aanderud, And Data Inc
Angela Hall, SAS
One of the major benefits of using SAS® Stored Processes is extendibility.
SAS® stored processes are one of the most customizable products; there are
several advantages, such as the ability to set up reports that can run in
various locations, enhance out-of-the box functionality with custom
widgets, and leverage all of the stored process server options. In this
discussion, you will learn advanced tips and tricks for using stored
processes within SAS BI clients.
5:30 p.m.
Escape from Big Data Restrictions by Leveraging
Advanced OLAP Cube Techniques
Stephen Overton, Overton Technologies LLC
Paper 047-2013
In today’s fast-growing field of business analytics, there are many tools and
methods for summarizing and analyzing big data. This paper focuses
specifically on OLAP technology and features natively available in OLAP
cubes that enable organizations to deploy robust business intelligence
reporting when high volumes of data exist. This paper discusses how to use
8
2:00 p.m.
What's New in SAS® Data Management
Nancy Rausch, SAS
Paper 070-2013
The latest releases of SAS® Data Integration Studio and SAS® Data
Management provide an integrated environment for managing and
transforming your data to meet new and increasingly complex data
management challenges. The enhancements help develop efficient
processes that can clean, standardize, transform, master, and manage your
data. Latest features include capabilities for building complex job
processes; new web-based development and job-monitoring
environments; enhanced ELT transformation capabilities; big data
transformation capabilities for Hadoop; integration with the SAS® LASR™
platform; enhanced features for lineage tracing and impact analysis; and
new features for master data and metadata management. This paper
provides an overview of the latest features of the products and includes use
cases and examples of the product capabilities.
3:00 p.m.
Bigger Data Analytics: Using SAS® on Aster Data and
Hadoop
John Cunningham, Teradata
Paper 071-2013
With the increased popularity of new Big Data clustered processing
platforms, SAS® Analytics now has the opportunity to solve newer, bigger
problems than ever before. Paper will focus on the evolution of Big Data
analytics, the new data sources and types, new technologies involved, to
achieve end to end analytic processing with SAS. Will specifically
demonstrate the use of new Big Data technologies, SAS Analytics with SAS/
ACCESS® for Aster, Aster SQL-MR, SQL-H to integrate end to end Big Data
analytics on the Aster Discovery Platform, even from raw data files stored
on Hadoop clusters.
3:30 p.m.
SAS-Oracle Options and Efficiency: What You Don't
Know Can Hurt You
John Bentley, Wells Fargo Bank
SAS/Access engines allow SAS to read, write, and alter almost any relational
database. Using the engine right out of the box works OK, but there are a
host of options that if properly used can improve performance, sometimes
greatly. In some cases though an incorrect option value will degrade
performance. This paper will review cross-database SAS/Access engine
options that can impact performance. Examples and test cases using an
Oracle database will be provided. All levels of SAS programmers, Enterprise
Guide users, and non-Oracle database users will find the paper useful.
4:30 p.m.
Data Mining and Text Analytics — Room 2004
SAS® Data Management Techniques: Cleaning and
Transforming Data for Delivery of Analytic Data Sets
3:00 p.m.
Chris Schacherer, Clinical Data Management Systems, LLC
Paper 540-2013
The analytic capabilities of SAS® software are unparalleled. Similarly, the
ability of the Output Delivery System to produce an endless array of
creative, high-quality reporting solutions is the envy of many analysts using
competing tools. Beneath the glitz and glitter is the dirty work of cleaning,
managing, and transforming raw source data and reliably delivering
analytic data sets that accurately reflect the processes being analyzed.
Although a basic understanding of DATA step processing and PROC SQL is
assumed, the present work provides examples of both basic data
management techniques for profiling data as well as transformation
techniques that are commonly utilized in the creation of analytic data
products. Examples of techniques for automating the generation and
delivery of production-quality, enterprise-level data sets are provided.
5:00 p.m.
How to Do a Successful MDM Project in SAP Using SAS®
MDM Advanced
Casper Pedersen, SAS Denmark
Paper 074-2013
How difficult would it be to embark on master data management (MDM)
projects at a large SAP organization? SAP organizations often come with
lots of opinionated people, ambitious project plans, and vast complexities
and politics. Don’t go with the Big Bang approach; instead, try a controlled
evolution. Come and see how one approach to the KNA1 (Customer Master)
table and MARA (Material Data) table was implemented.
2:00 p.m.
Using Data Mining in Forecasting Problems
Timothy Rey, Dow Chemical Company
Chip Wells, SAS
Justin Kauhl, Tata Consulting
In today’s ever-changing economic environment there is ample
opportunity to leverage the numerous sources of time series data now
readily available to the savvy business decision maker. Time series data can
be used for business gain if the data is converted to information and then
knowledge. Data mining processes, methods, and technology oriented to
transactional-type data (data not having a time series framework) have
grown immensely in the last quarter century. There is significant value in
the interdisciplinary notion of data mining for forecasting when used to
solve time series problems. The presentation describes how to get the most
value out of the myriad of available time series data by utilizing data mining
techniques specifically oriented to data collected over time.
Time Is Precious, So Are Your Models: SAS® Provides
Solutions to Streamline Deployment
Jonathan Wexler, SAS
Wayne Thompson, SAS
Paper 086-2013
Organizations spend a significant amount of time, often too much,
operationalizing models. The more time you can spend on analytics, and
the less time on deployment headaches, the better chance you have to
address core business challenges. This paper shows you how SAS has
accelerated data mining model deployment throughout the analytical life
cycle, by providing key integration points across SAS® solutions. Whether
you built your models using SAS® High-Performance Analytics Server, SAS®
Enterprise Miner™, or SAS/STAT®, this paper shows you how to automate
the management, publishing, and scoring of models by using SAS® Data
Integration Studio, SAS® Model Manager, and SAS® Scoring Accelerator.
Immediate benefits include reduced data movement, increased
productivity of analytic and IT teams, and faster time to results.
4:00 p.m.
Bringing Churn Modeling Straight to the Source: SAS®
and Teradata In-Database Model Development
Karl Krycha, Teradata
Jonathan Wexler, SAS
Paper 087-2013
This paper takes a closer look at the opportunities of using the predictive
analytic power of SAS® together with the performance and scalability of
Teradata. Users will see how the SAS® Analytics Accelerator for Teradata
improves modeling speed from hours to seconds, allowing users to
produce more models faster. The SAS Analytics Accelerator eliminates data
movement by moving SAS analytic computations capabilities to the
Teradata database. The paper provides an overview of the available
procedures and uses a typical business application to illustrate the full endto-end process of analytic modeling within Teradata.
4:30 p.m.
Demand Forecasting Using a Growth Model and
Negative Binomial Regression Framework
Michelle Cheong, Singapore Management University
Cally Ong Yeru, Singapore Management University
Murphy Choy, Singapore Management University
Paper 088-2013
In this paper, we look at demand forecasting by using a growth model and
negative binomial regression framework. Using cumulative sales, we model
the sales data for different wristwatch brands and relate it to their sales and
growth characteristics. We apply clustering to determine the distinctive
characteristics of each individual cluster. Four different growth models are
applied to the clusters to find the most suitable growth model to be used.
After determining the appropriate growth model to be applied, we then
forecast the sales by applying the model to new products being launched
in the market and continue to monitor the model further.
9
5:00 p.m.
11:30 a.m.
Using Classification and Regression Trees (CART) in SAS®
Enterprise Miner™ for Applications in Public Health
Double-Clicking a SAS® File: What Happens Next?
Paper 089-2013
Paper 115-2013
Leonard Gordon, University of Kentucky
Classification and regression trees (CART)—a non-parametric methodology
—were first introduced by Breiman and colleagues in 1984. In this paper
they are employed using SAS® Enterprise Miner™, and several examples are
given to demonstrate their use. CART are underused (especially in public
health), and they have the ability to divide populations into meaningful
subgroups that allow the identification of groups of interest and enhance
the provision of products and services accordingly. They can provide a
simple yet powerful analysis. This paper attempts to demonstrate their
value and thus encourage their increased use in data analysis.
Sandy Gibbs, SAS
Donna Bennett, SAS
5:30 p.m.
When you double-click a SAS® file on your desktop or in Windows Explorer,
which program launches—SAS or SAS® Enterprise Guide®? Does the SAS
program open in a preferred application? Both Microsoft and SAS have
introduced changes in how you change the default SAS file type
associations, starting with Microsoft Windows Vista and with SAS 9.2. If you
have installed SAS 9.2 and are using earlier approaches to change the
default file associations (for example, through Windows Explorer), you
might encounter problems. This topic has generated a lively discussion
among the SAS Deployment Support Community! The recommended
method for changing file type associations is described in this paper. Also
discussed are ways to troubleshoot problems that might surface during this
process.
Opinion Mining and Geo-positioning of Textual
Feedback from Professional Drivers
2:00 p.m.
Mantosh Kumar Sarkar, Oklahoma State University
Goutam Chakraborty, Oklahoma State University
Paper 500-2013
While many companies collect feedback from their customers via mobile
applications, they often restrict their analysis to numeric data and ignore
analyzing customer feedback and sentiments from textual data. In this
paper, we analyze customer feedback by professional drivers sent via a
mobile app. We demonstrate how SAS® Text Miner can be used to
automatically generate and summarize topics from positive and negative
feedbacks. In addition, we demonstrate how SAS® Sentiment Analysis
studio can be used to build rules to predict customers’ sentiments
automatically so that experts’ time can be used for more strategic purposes.
Finally, we show how feedback with positive and negative sentiments can
be geo-positioned on the U.S. map via JMP® scripts to provide a better
visualization of sentiment distribution.
Foundations and Fundamentals — Room 2008
10:30 a.m.
Quick Hits: My Favorite SAS® Tricks
Marje Fecht, Prowerk Consulting
Paper 114-2013
Are you time-poor and code-heavy? It's easy to get into a rut with your SAS®
code, and it can be time-consuming to spend your time learning and
implementing improved techniques. This presentation is designed to share
quick improvements that take 5 minutes to learn and about the same time
to implement. The quick hits are applicable across versions of SAS and
require only Base SAS® knowledge. Included are: - little-known functions
that get rid of messy coding - simple macro tricks - dynamic conditional
logic - data summarization tips to reduce data and processing - generation
data sets to improve data access and rollback - testing tips
A Day in the Life of Data - Part 1
Brian Bee, The Knowledge Warehouse Ltd
As a new SAS® programmer, you may be overwhelmed with the variety of
tricks and techniques that you see from experienced SAS programmers; as
you try to piece together some of these techniques you get frustrated and
perhaps confused because the data showing these techniques are
inconsistent. That is, you read several papers and each uses different data.
This series of four papers is different. They will step you through several
techniques but all four papers will be using the same data. The authors will
show how value is added to the data at each of the four major steps: Input,
Data Manipulation, Data and Program Management, and Graphics and
Reporting.
3:00 p.m.
Harry Droogendyk, Stratia Consulting Inc.
Reporting.
4:00 p.m.
Peter Crawford, Crawford Software Consultancy Limited
10
Reporting.
5:00 p.m.
A Day in the Life of Data: Part 4 - Graphics and Reporting
Sanjay Matange, SAS
Paper 119-2013
Although we often work in our own department with little contact with
others, whether they’re on the next floor or halfway around the world,
everyone has an impact on each other. In this four-part series, each paper
examines one of the four major aspects of SAS® Data Management: initial
data input; data manipulation; data and program management; and
graphics and reporting. Each paper teaches some fundamental skills and
shows how each step adds value to the data.
Hands-on Workshops — Room 2011
10:30 a.m.
Adding New Rows in the ADaM Basic Data Structure:
When and How
Mario Widel, Roche Molecular Systems
Sandra Minjoe, Octagon Research Solutions
The ADaM (Analysis Data Model) BDS (Basic Data Structure) has specific
rules to follow when adding columns or rows. Because there are limitations
to what can be added as a column, much of our derived content must be
added as rows. This HOW uses a Vital Signs example, demonstrating the
common BDS need of adding analysis parameters and visits. Attendees will
use a general specification and mock-up to create metadata content that
can be used for both a detailed specification and within a define document.
The resulting content will include variable-level metadata, parameter-level
metadata, and SAS® code snippets. This is an intermediate-level HOW.
Attendees are expected to be familiar with the analysis needs of clinical
trials, CDISC, and submissions to FDA.
10:30 a.m.
SAS® Workshop: SAS® Add-In for Microsoft Office 5.1
11:30 a.m.
Paper 520-2013
SAS® Workshop: Creating SAS® Stored Processes
Eric Rossland, SAS
This workshop provides hands-on experience using the SAS® Add-In for
Microsoft Office. Workshop participants will:
• access and analyze data
• create reports
• use the SAS add-in Quick Start Tools
Eric Rossland, SAS
Paper 521-2013
This workshop provides hands-on experience creating SAS® Stored
Processes. Workshop participants will:
• use SAS® Enterprise Guide® to access and analyze data
• create stored processes which can be shared across the organization
10:30 a.m.
The Armchair Quarterback: Writing SAS® Code for the
Perfect Pivot (Table, That Is)
“Can I have that in Excel?” This is a request that makes many of us shudder.
Now your boss has discovered Microsoft Excel pivot tables. Unfortunately,
he has not discovered how to make them. So you get to extract the data,
massage the data, put the data into Excel, and then spend hours rebuilding
pivot tables every time the corporate data are refreshed. In this workshop,
you learn to be the armchair quarterback and build pivot tables without
leaving the comfort of your SAS® environment. You learn the basics of Excel
pivot tables and, through a series of exercises, how to augment basic pivot
tables first in Excel, and then using SAS. No prior knowledge of Excel pivot
tables is required.
• access the new stored process from the SAS® Add-In for Microsoft Office
2:00 p.m.
SAS® Workshop: SAS® Data Integration Studio Basics
Kari Richardson, SAS
Paper 522-2013
This workshop provides hands-on experience using SAS Data Integration
Studio to construct tables for a data warehouse. Workshop participants will:
• define and access source data
• define and load target data
• work with basic data cleansing
11
2:00 p.m.
SAS® Enterprise Guide® 5.1: A Powerful Environment for
Programmers, Too!
Marje Fecht, Prowerk Consulting
Rupinder Dhillon, Dhillon Consulting Inc
Have you been programming in SAS® for a while and just aren't sure how
SAS® Enterprise Guide® can help you? This presentation demonstrates how
SAS programmers can use SAS Enterprise Guide 5.1 as their primary
interface to SAS, while maintaining the flexibility of writing their own
customized code. We explore:
• navigating and customizing the SAS Enterprise Guide environment
• using SAS Enterprise Guide to access existing programs and enhance
processing
• exploiting the enhanced development environment including syntax
completion and built-in function help
• using SAS® Code Analyzer, Report Builder, and Document Builder
• adding Project Parameters to generalize the usability of programs and
processes
• leveraging built-in capabilities available in SAS Enterprise Guide to
further enhance the information you deliver
• Review / create a SAS Data Integration Studio job that will execute the
uploaded data jobs on the DataFlux Data Management Server
4:00 p.m.
SAS® Workshop: SAS® Visual Analytics 6.1
Eric Rossland, SAS
Paper 524-2013
This workshop provides hands-on experience with SAS® Visual Analytics.
Workshop participants will:
• explore data with SAS® Visual Analytics Explorer
• design reports with SAS® Visual Analytics Designer
5:00 p.m.
SAS® Workshop: DataFlux® Data Management Studio
Basics
Paper 525-2013
This workshop provides hands-on experience using DataFlux® Data
Management Studio to profile then cleanse data. Workshop participants
will:
• learn to navigate DataFlux® Data Management Studio
• define and run a data profile
• define and run a data job
2:00 p.m.
IT Management — Salon 10,11, 12
Using PROC FCMP to the Fullest: Getting Started and
Doing More
1:30 p.m.
The FCMP procedure is used to create user-defined functions. Many users
have yet to tackle this fairly new procedure, while others have attempted to
use only its simplest options. As with many tools within SAS®, the true value
of this procedure is appreciated only after the user has started to learn and
use it. The basics can quickly be mastered, and this allows the user to move
forward to explore some of the more interesting and powerful aspects of
the FCMP procedure. The use of PROC FCMP should not be limited to the
advanced SAS user. Even those fairly new to SAS should be able to
appreciate the value of user-defined functions.
Manage Your Data as a Strategic Asset
Khaled Ghadban, Canada Post
Richard Beaver, United Natural Foods, Inc
Bill Ford, Vail Resorts
3:00 p.m.
Searching for Business Value in Big Data with Hadoop
Mike Olson, Cloudera
Paul Kent, SAS
Gavin Day, SAS
3:00 p.m.
While some well-resourced organizations can simply throw technical talent
at uncovering the value in their big data, others struggle to find analytic
technology that takes full advantage of the richness and scale of the
Hadoop ecosystem. Join industry thought leaders from Cloudera, Intel and
SAS for a discussion of how the Hadoop community is using analytics to
derive critical insights that drive significant business impact from their big
data assets
SAS® Workshop: SAS® Data Integration Studio Advanced
Paper 523-2013
This workshop provides hands-on experience using a combination of
DataFlux Data Management Studio and SAS® Data Integration Studio.
• Review two DataFlux Data Management Studio data jobs
• Upload the DataFlux Data Management Studio data jobs to the DataFlux
Data Management Server
12
4:00 p.m.
3:30 p.m.
How IT Completes the Big Data Puzzle with Hadoop
Is the Legend in Your SAS/Graph® Output Still Telling the
Right Story?
Mike Olson, Cloudera
Pauline Nist, Intel
Paul Kent, SAS
Gavin Day, SAS
It's easy to become overwhelmed by the increasing volume, velocity and
variety of big data – and miss the value that it holds to uncovering
profitable insights and answering complex questions. So what's the missing
piece to solving the big data puzzle? Hadoop. IT organizations are rapidly
leveraging Hadoop to quickly derive a more complete picture and analysis
of all their data. Before you can get value from your data, it has to be well
organized, managed and governed. Thought leaders from Cloudera, Intel
and SAS will share key insights on how IT can solve the big data puzzle with
Hadoop
Pharma and Health Care — Room 2000
2:00 p.m.
Automated and Customized Reports as a Single Image
File Using Graph Template Language (GTL): A Case Study
of Benchmarking Reports in Medical Research
Monarch Shah, ICON Clinical Research
Ginny Lai, ICON Late Phase & Outcomes Research
Eric Elkin, ICON
Paper 495-2013
Site benchmarking reports give us an overview of demographic, clinical,
and disease characteristics for the individual site with comparison to the
study as a whole. A solution was needed for on-going reporting to over 250
study sites. The report needed to be concise and present data in both
tables and figures. (This objective could also arise, for example, in
comparing each store’s performance to the entire chain or each classroom’s
performance to the school district.) However, creating and combining
tables and figures into a document can be challenging. Graph Template
Language (GTL) provides a powerful alternative to customize and automate
benchmarking reports. This paper will focus on using GTL to create panels
comprised of descriptive tables and multiple graphs into a single image file.
2:30 p.m.
Patient Profile Graphs Using SAS®
Sanjay Matange, SAS
Paper 160-2013
Patient profiles provide information on a specific subject participating in a
study. The report includes relevant data for a subject that can help correlate
adverse events to concomitant medications and other significant events as
a narrative or a visual report. This presentation covers the creation of the
graphs useful for visual reports from CDISC data. It includes a graph of the
adverse events by time and severity, graphs of concomitant medications,
vital signs and labs. All the graphs are plotted on a uniform timeline, so the
adverse events can be correlated correctly with the concomitant
medications, vital signs and labs. These graphs can be easily incorporated
with the rest of the demographic and personal data of the individual
patient in a report.
Alice Cheng, Chiltern Inc.
Justina Flavin, self employed
In clinical studies, researchers are often interested in the effect of treatment
over time for multiple treatments or dosing groups. Usually, in a graphical
report, the measurement of treatment effect is on the vertical axis and a
second factor, such as time or visit, on the horizontal axis. Multiple lines are
displayed in the same figure; each line represents a third factor, such as
treatment or dosing group. It is critical that the line appearance (color,
symbol and style) is consistent throughout the entire clinical report as well
as across clinical reports from related studies.
4:30 p.m.
Variance Partition: My Mission and Ambition Come to
Fruition
Brenda Beaty, University of Colorado
L. Miriam Dickinson, University of Colorado
Paper 162-2013
In medical research, we are often interested in understanding the complex
interplay of variables with one or more clinical outcomes. Because our
bodies are always in motion, simply viewing a snapshot of data in time is
sub-optimal. Longitudinal data gives us the advantage of modeling 'reallife' time-dependent variables and outcomes. This paper is an exploration
of one such project. In this paper, we first familiarize ourselves with a study
of the relationship of diabetic nephropathy and blood pressure measured
longitudinally. We then explore a number of ways to model the data, with
the final goal of using time-varying covariates to model the illness path, as
well as the ultimate outcome, thereby getting complete partitioning of the
variance.
5:00 p.m.
Quantile Regression in Pharmaceutical Marketing
Research
George Mu, IMS health Inc
Paper 163-2013
In pharmaceutical marketing research, the heterogeneity in healthcare data
presents lots of challenges to researchers. Managers have a difficult time
getting comprehensive market pictures from simple equations that
generally fit all individuals. Quantile regression offers an efficient and
robust way to tease out the different patterns existing in the healthcare
market. This paper demonstrates the value of applying quantile regression
to solve pharmaceutical marketing research problems. We illustrate the
methodology by using SAS® QUANTREG and QUANTLIFE procedures to
compare physicians’ new product uptake patterns; to find influential drivers
in patient medication compliance; and to help in the design of clinical trials
for patient selections. The results from these empirical examples show
quantile regressions provide more market insights than other commonly
used methodologies.
13
5:30 p.m.
Using PROC GENMOD to Investigate Drug Interactions:
Beta Blockers and Beta Agonists and Their Effect on
Hospital Admissions
Hui Fen Tan, Columbia University
Ronald Low, New York City Health and Hospitals
Corporation
Shunsuke Ito, New York City Health and Hospitals
Corporation
Raymond Gregory, New York City Health and Hospitals
Corporation
Vann Dunn, New York City Health and Hospitals Corporation
for a group of this nature to be successful in national and global
organizations. It also reviews technologies that could be beneficial for
bridging the communication gap in a user group of this type.
Planning and Support — Room 3016
3:00 p.m.
Branding Yourself Online
Kirsten Hamstra, SAS
Shelly Goodin, SAS
Paper 184-2013
Every year, more than half a million adverse reactions to drugs are reported
to the FDA. This paper is a real-world, large-scale review of beta blockers
and beta agonist usage. We use New York City public hospitals’ records to
investigate whether interactions of beta blockers and beta agonists are
associated with adverse medical outcomes such as increased hospital visits,
a common indicator of health care quality. The GENMOD procedure in SAS®
provides a variety of count data models, including Poisson regression and
negative binomial regression. We find that patients on “non-clinical trials
use” of beta blockers and beta agonists, older patients, and patients with
history of COPD, CAD, and pneumonia tend to have higher hospital visit
rates.
Your online reputation matters. Whether you’re using social media for
professional or personal reasons, it’s important to understand and control
your public persona. For those looking to further your career, build your
business, or enter the workforce, you can maximize your positive exposure
by knowing where and how to engage online. This presentation will
highlight some of the best practices for online engagement, provide
suggestions for where to engage, and showcase some examples. To anyone
engaging in social media who wants to better understand the impact of
their contributions and control their online reputation, this presentation is
for you. Takeaways: · Craft an incredible bio · Harness the power of SEO ·
Strengthen your online reputation · Engage—where and how to do it ·
Make meaningful connections Audience: · Job seekers · Consultants ·
Students
2:00 p.m.
4:00 p.m.
SAS® Skill Learning and Certification Preparation in a
Graduate School Setting
Coaching SAS® Beginner Programmers: Common
Problems and Some Solutions
Paper 164-2013
Christine Bonney, University of Pennsylvania
Michael Keith, Jr., University of Pennsylvania
Paper 182-2013
A semester-long course was created with the goal of teaching graduate
students SAS® programming skills and to prepare them to take the SAS®
Certified Base Programming for SAS®9 exam. Course activities and materials
include: weekly lectures; in-class labs; take-home problem sets; virtual
(online) labs; assigned readings from the “SAS® Certification Prep Guide:
Base Programming for SAS®9”; midterm and final exams; and access to SAS®
OnDemand for Academics. This paper covers the details of the course
development and design, as well as preliminary results from the course and
plans for future developments.
2:30 p.m.
Considerations for Creating an In-House SAS® User
Group in a Geographically Disbursed Organization
Stefanie Reay, Qualex Consulting Services, Inc.
Paper 183-2013
This presentation will review considerations for creating in-house SAS® user
groups in geographically disbursed organizations, in which in-person user
group meetings are not cost-effective or not feasible, but for which an inhouse SAS user group would still be beneficial. It defines in-house SAS user
groups, and overviews the resources available from SAS for starting and
continuing an in-house SAS user group. It discusses benefits and challenges
of starting/maintaining an in-house SAS user group, options for
organizational structures of in-house SAS user groups, and unique needs
14
Peter Timusk, Statistics Canada
Paper 185-2013
This paper will present a number of problems SAS® beginner programmers
encounter when first writing SAS programs. The paper will cover three
cases and show how pointing out patterns to beginner programmers will
aid them in avoiding errors in their SAS code.
4:30 p.m.
A CareerView Mirror: Another Perspective on Your Work
and Career Planning
Bill Donovan, Ockham Source
Paper 186-2013
Career planning in today’s tumultuous job market place requires a more
rigorous and disciplined approach that must begin with each individual
tracking and evaluating distinctive skills and experiences. With an emphasis
on the SAS® professional and the career track unique to the programmers'
challenges, this paper is designed to set the stage for professional reflection
and career planning. The ability to organize and inventory your entire
career-related experiences is the foundation of a solid plan. The catalog of
your work assignments and functional responsibilities creates a reflection
of your efforts in your career to date. All of this helps to build your
CareerView Mirror, which provides another perspective on your work and
career planning.
5:00 p.m.
2:00 p.m.
Gotchas: Hidden Workplace and Career Traps to Avoid
SAS® Essentials: Maximize the Efficiency of Your Most
Basic Users
Steve Noga, Rho
Paper 187-2013
Being successful at your job takes more than just completing your tasks
accurately and on time. There are hidden holes everywhere, some deeper
than others, that must be navigated; yet no map exists for you to follow.
Most companies have a set of stated policies or rules that their employees
are expected to follow, but what about the unstated ones that may have an
effect on how fast or how far you advance within the company? Hidden
traps also exist along the way of your career path. This panel discussion will
highlight some “gotchas” of which you should be aware and ways to keep
from falling into the holes.
Poster and Video Presentations — SAS Support
and Demo Area
2:00 p.m.
Create a Nomogram with SGPlot
Julie Kezik, Yale University
Melissa Hill, Yale University
Paper 196-2013
If programming and research assistants were taught SAS® essentials, job
efficiency could be maximized with the ability to use SAS as a tool to do
their own preparatory work for assigned tasks. This paper summarizes a
supplemental training program which teaches basic SAS programming
skills to enable support staff to be more independent.
2:00 p.m.
Using CALL SYMPUT to Generate Dynamic Columns in
Reports
Sai Ma, pharmanet-i3
Suwen Li, Everest Clinical Research Services, Inc.
Regan Li, Hoffmann-La Roche Limited
Bob Lan, Everest Clinical Research Services, Inc.
Cynthia Loman, Genomic Health Inc
Paper 198-2013
The nomogram that I created shows the relative values of four predictors
from a logistic model along with a line showing cumulative model score
and cumulative model risk. I have programmed it with PROC SGPLOT and
used natural splines for one of the variables in my model. I used prostate
cancer data for my example.
When creating reports, we often want to make the report respond
dynamically to data. If the headers and the number of columns in the report
are unknown, it is helpful when they change dynamically depending on the
data. As a powerful SAS® procedure, PROC TABULATE can produce dynamic
results in most cases. This paper describes how to use the CALL SYMPUT
routine and PROC REPORT to generate dynamic columns in reports in cases
where PROC TABULATE does not yield the desired results.
Paper 194-2013
2:00 p.m.
Using SAS to Create Code for Current Triage Systems
during Chemical Incidents
Abbas Tavakoli, University of South Carolina
Erik Svendsen, University of tulane
Jean Craig, MUSC
Joan Culley, University of South Carolina
2:00 p.m.
From SDTM to ADaM
Sai Ma, pharmanet-i3
Suwen Li, Everest Clinical Research Services, Inc.
Regan Li, Hoffmann-La Roche Limited
Bob Lan, Everest Clinical Research Services, Inc.
Paper 195-2013
Paper 199-2013
Chemical incidents involving irritant chemicals such as chlorine pose a
significant threat to life and require rapid assessment. This paper used the
first outcomes-level study (R21 NIH) involving an actual mass casualty
chemical incident to create code for four triage systems (CBRN, SALT,
START, and ESI). Data used for this paper, which come from six datasets
collected by the project team from a 60-ton railroad chlorine leak in 2005 in
Graniteville, South Carolina, include patient demographics, exposure
estimates, symptoms, outcome categories, and physiological
measurements. Data collected for approximately 900 victims of the chlorine
leak were merged to generate a research dataset. SAS® 9.2 was used to
create code from logic to mimic the triage decision tree, yielding
classifications for each system.
The use of SDTM and ADaM standards are highly desirable in FDA
guidances. More and more sponsors submit both of these standards to
regulatory authorities. When SDTM data sets are more common, ADaM is
usually derived from SDTM. However, the SDTM distinctive data structure
causes problems when deriving ADaM data. This paper describes problems
encountered when deriving ADaM data and provides resolutions and
examples.
2:00 p.m.
Exploring the PROC SQL _METHOD Option
Charlie Shipp, Consider Consulting, Inc.
Paper 200-2013
The SQL procedure has powerful options for users to take advantage of.
This presentation explores the fully supported _METHOD option as an
applications development and tuning tool. Attendees learn how to use this
powerful option to better understand and control how a query processes.
15
2:00 p.m.
2:00 p.m.
A Practical Approach to Creating Define.XML by Using
SDTM Specifications and Excel functions
A Simple Macro to Minimize Data Set Size
Paper 201-2013
Define.xml (Case Report Tabulation Data Definition Specification) is a part
of new drug submission required by the FDA. Clinical SAS® programmers
usually use SAS programming [1, 2, 3, 4, 5] to generate the code of
Define.xml as described in the CDISC Case Report Tabulation Data
Definition Specification (define.xml) V1.0.0 [6]. This paper illustrates the
process of using SDTM specifications and Excel functions to generate the
code of Define.xml in an easy and straightforward way.
Whenever you submit either SDTM or ADaM data sets to FDA, if any SAS®
data set is great than 1 GB in size, FDA will ask you to split the data set. In
fact, since the length of a variable affects both the amount of disk space
used and the number of I/O operations required to read and write the data
set, resizing text columns to fit the longest value within the column is
applicable to every field that uses SAS data sets in their business. To help
save resources and improve data mining efficiency, this paper discusses a
simple macro to minimize the size of a SAS data set.
2:00 p.m.
2:00 p.m.
Extending the Power of Base SAS® with Microsoft Excel
Basel II Advanced IRB in Commercial Banking: Quantify
the Borrower and Guarantor by Two-Step Scoring Model
Amos Shu, Endo Pharmaceuticals
Shilpa Khambhati, Mathematca Policy Research Inc.
Paper 203-2013
The SAS Macro Language is an invaluable SAS tool that can be used for
iterative SAS data processing, eliminating redundancy in SAS code. Using
the SAS Macro Language with Microsoft Excel makes programming tasks
even easier. This paper describes using the SAS Macro Language and
Microsoft Excel to automatically generate customized reports. The
proposed method uses Excel macros to drive SAS macros without having to
open SAS programs and manually upgrade parameters specific for each
site’s data when the data becomes available. The process eliminates
manually editing SAS programs and improves data quality by reducing
programming error and program maintenance time.
2:00 p.m.
Using LinkedIn to Find Your Next SAS® Job
Paper 204-2013
LinkedIn is fast becoming a great place for SAS® recruiters and SAS
candidates to meet. If you are looking for a job, this poster provides some
tips to spiff up your LinkedIn profile to get the SAS programming job of
your dreams.
2:00 p.m.
Selection Group Prompts with SAS® Stored Processes:
More Power, Less Programming
Angela Hall, SAS
Paper 205-2013
Many programmers either do not know or understand how to use the
selection group prompts to make advanced stored processes a little easier
to manage. Many times, end users have these crazy requirements and a
programmer can use the selection group prompt instead of writing 10
different stored processes.
16
Paper 206-2013
Hengrui Qu, Citi Group Inc.
juan zhao, Citi Group Inc
Paper 207-2013
For public companies, the probability of default usually adopts well-known
structural and reduced form credit risk models. However, in commercial
lending, there are large portfolios of unlisted companies, which could not
use these two approaches. Furthermore, privately held companies
commonly get a guarantor to enhance their credibility during loan
application. Unlike the single logistic model used for retail credit risk
analysis, two- step credit scoring models could be used to quantify both
borrower and guarantor's risk exposed to unlisted companies based on the
limited information maximum likelihood. This paper will focus on how to
quantify the risk for commercial borrowers with guaranty by two-step
scoring model, which provides Basel II advanced IRB risk measure: the PD
for the commercial customer and transaction.
2:00 p.m.
A Unique Approach to Create Custom Reports By
Leveraging the Strengths of SAS® and Excel
Amy Overby Wilkerson, RTI International
Brett Anderson, RTI International
Barbara Bibb, RTI International
Mai Nguyen, RTI International
Paper 208-2013
Survey projects often require custom reports to allow project staff to
monitor production as well as various statistics from the collected data. At
RTI, we've come up with a unique approach for creating custom reports for
our projects by leveraging the strengths of SAS® and Excel. In SAS, we use
PROC SQL to select and when necessary aggregate data. After processing
the data in SAS, results are sent to Excel for reporting and graphics. In our
paper, we will present a few sample reports, program codes and the
detailed explanations of how these reports were created.
2:00 p.m.
Working with a Large Pharmacy Database: Hash and
Conquer.
David Izrael, Abt Associates
Paper 209-2013
Working with a large pharmacy database means having to process - merge,
sort, and summarize - hundreds of millions of observations. By themselves,
traditional methods of processing can lead to prohibitive data processing
times that endanger deadlines. The hash object is the fastest and most
versatile method in the SAS® system of substantially accelerating the
processing. In our paper, we apply hash methods to a routine lookup
function where one needs to merge the kernel pharmacy database with its
satellites. We also present comparatively new nontraditional features of the
hash object, such as handling duplicate keys and finding frequency
counters. At the same time, we underscore the necessity of traditional sortand-merge methods, but suggest that they be used carefully.
2:00 p.m.
With a Trace: Making Procedural Output and ODS
Output Objects Work for You
Louise Hadden, Abt Associates Inc.
The Output Delivery System (ODS) delivers what used to be printed output
in many convenient forms. What most of us don't realize is that "printed
output" from procedures (whether the destination is PDF, RTF, or HTML) is
the result of SAS® packaging a collection of items that come out of a
procedure that most people want to see in a predefined order (aka
template). With tools such as ODS TRACE, PROC CONTENTS, and PROC
PRINT, this paper explores the many buried treasures of procedural output
and ODS output objects and demonstrates how to use these objects to get
exactly the information that is needed, in exactly the format wanted.
2:00 p.m.
Analyzing the Safewalk Program with SAS®: Saving
Shelter Dogs One Walk at a Time
Terri Bright, MSPCA Boston
footnotes, ODS text fields and tabular output; and add custom "fills" to SAS/
GRAPH® maps and graphs. Some possible uses of custom images include a
company logo embedded in SAS output, graphic displays of positive or
negative outcomes, and watermarks containing "draft" or "confidential".
The SAS code to accomplish all these potential uses, and more, will be
shown.
2:00 p.m.
Weighted Sequential Hot Deck Imputation: SAS® Macro
vs. the SUDAAN PROC HOTDECK
David Izrael, Abt Associates
Michael Battaglia, Abt Associates Inc
Paper 213-2013
Item non-response is a challenge faced by all surveys. Item non-response
occurs when a respondent skips over a question, refuses to answer a
question, or does not know the answer to a question. Hot deck imputation
is one of the primary imputation tools used by survey statisticians. Recently,
a new competitor in the field of Weighted Sequential Hotdeck Imputation
has arrived: PROC HOTDECK of SUDAAN, version 10. We compared the
results of imputation using the new procedure with the results of the
Hotdeck SAS® Macro with respect to: a) how close the post-imputation
weighted distributions and standard errors of the estimates are to those of
the item respondent data; b) whether there is a difference in the number of
times donors contribute to the imputation.
2:00 p.m.
Creating ZIP Code-Level Maps with SAS®
Barbara Okerson, WellPoint
SAS®, SAS/GRAPH®, and ODS graphics provide SAS programmers with the
tools to create professional and colorful maps. Provided with SAS/GRAPH
are boundary files for U.S. states and territories, as well as internal
boundaries at the county level. While much data and results can be
displayed at this level, often a higher degree of granularity is needed. The
U.S. Census Bureau provides ZIP code boundary files in ESRI shape file
format (.shp) by state for free download and import into SAS using SAS
PROC DATAIMPORT. This paper illustrates the use of these ZIP code
tabulation area (ZCTA) files with SAS to map data at a ZIP code level.
Example maps include choropleth, distance, and heat maps.
The MSPCA in Boston initiated the Safewalk Program in January 2009. This
program was designed to enrich the experience of shelter dogs by
providing training to volunteers and staff that allow dogs of varied
backgrounds and temperaments to be exercised safely, as well as
promoting behaviors encouraging adoption on the adoption floor. A data
extract from the MSPCA's Chameleon data base was analyzed using
multiple SAS® procedures in SAS/STAT®. This paper will demonstrate how
SAS analysis, output, and statistical graphs allowed us to assess the effects
of the Safewalk Program and which populations it most affected.
2:00 p.m.
Behind the Scenes with SAS®: Using Custom Graphics in
SAS Output
SAS® provides many opportunities to add customized images to SAS ODS
output. This presentation will demonstrate various ways to add custom
backgrounds to tabular and graphic output; add custom images to titles,
2:00 p.m.
A Case Application of Propensity Score Matching in MTM
Outcomes Evaluation at Retail Pharmacy
Michael Taitel, Walgreens
Zhongwen Huang, Walgreens
Youbei Lou, Walgreens
Paper 215-2013
Propensity score matching approaches in outcomes analysis are often used
to reduce the potential bias in observational studies. The process includes
propensity score estimation, matching and evaluation. This paper presents
a case application in Outcome Evaluation of Medication Therapy
Management at Retail Pharmacy. Baseline outcome metrics, which do not
appear in the propensity score estimation model, were checked for balance
later to detect if there are any important covariates that affect both
treatment and outcomes have been neglected. In addition, by
appropriately selecting variables in retail pharmacy environment, one
matching for multiple outcomes analysis, which works as a pseudo
randomization study, can improve efficiency.
17
2:00 p.m.
2:00 p.m.
Using SAS® to Expand the Application of Standard
Measures and Guide Statistical Explorations: Creating
Healthy Eating Index Scores Using Nutrition Data
System for Research Output
Propensity Score-Based Analysis of Short-Term
Complications in Patients with Lumbar Discectomy in
the ACS-NSQIP Database
David Ludwig, University of Miami
David Landy, University of MIami Miller School of Medicine
Joy Kurtz, Univ. of Miami
Tracie Miller, University of Miami
Paper 216-2013
We created a SAS® program to calculate a measure of diet quality, the
Healthy Eating Index (HEI, http://www.cnpp.usda.gov/
HealthyEatingIndex.htm), using output from a widely applied dietary
software package, Nutrition Data System for Research (NDSR, http://
www.ncc.umn.edu/products/ndsr.html). Currently, application of the HEI in
research and clinical assessment is limited by the challenges posed in
calculating the HEI using the highly complex and detailed NDSR output.
The SAS program extracts the required NDSR output files and then
calculates the combination of algebraic manipulations and logical
statements to obtain HEI scores. We also offer suggestions for increasing
usability, such as with the %INCLUDE statement, and show how the
program can be used to explore related statistical issues such as reliability
via PROC MIXED.
2:00 p.m.
A SAS® Macro for Generating a Set of All Possible
Samples with Unequal Probabilities without
Replacement
Alan Silva, Universidade de Brasilia
Paper 217-2013
This paper considers listing all possible samples of size n with unequal
probabilities without replacement in order to find the sample distribution.
The main application of that is to estimate the Horvitz-Thompson (HT)
estimator and possibly to know the shape of its sample distribution to
construct confidence intervals. The algorithm computes all possible
samples of the population, in contrast with PROC SURVEYSELECT which
generates any samples of size n, but not all possible samples, and at the
end it is possible to plot the sample distribution of the estimator. The
equations are encoded in a SAS/IML® macro and the graphics are made
using PROC GPLOT.
2:00 p.m.
Here Is How We Do It: Teaching SAS® at Community
Colleges
Meili Xu, Ohlone College
Paper 218-2013
Data is everywhere today, and SAS® programming skills are in high
demand. Providing community college students with SAS skills is extremely
valuable in preparing them for real-world job positions right after taking
the classes. In this paper, we will describe our experience and approaches
to teaching SAS to our students at Ohlone College. With the paper
presentation at the conference, we wish to instigate a dialogue among
other educators teaching SAS to share ideas and resources so that we may
all better equip students with strong SAS skills that will serve them well in
their future careers. At the workshop, you may also have an opportunity to
gain some hands-on experience on basic SAS procedures.
18
Yubo Gao, University of Iowa
Paper 220-2013
Lumbar discectomy is the most common spinal procedure performed, and
it can be done on an outpatient basis. In this study, we want to compare the
incidence of complications in patients undergoing single-level lumbar
discectomy between the inpatient and outpatient settings, to determine
baseline 30-day complication rates, and to identify independent risk factors
for complications. To achieve those, patients undergoing lumbar
discectomy between 2005 and 2010 were selected from the ACS-NSQIP
database, based on a single primary CPT code. Thirty-day post-operative
complications and pre-operative patient characteristics were identified and
compared. Propensity score matching and multivariate logistic regression
analysis were used to adjust for selection bias and identify predictors of 30day morbidity. All analyses are performed via SAS® software.
2:00 p.m.
Recovering SAS® User Group Proceedings for the SAS®
Community
Lex Jansen, lexjansen.com
Richard La Valley, Strategic Technology Solutions
Paper 221-2013
For many years, SAS® User Groups held conferences whose proceedings
were available only in print and only to those who attended or those who
knew that copies existed in the SAS Library in Cary, NC. Over the past
couple of years, there has been a project to digitize the printed proceedings
of SAS User Groups International, SAS European Users Group International,
NorthEast SAS Users Group, SouthEast SAS Users Group, Western Users of
SAS Software, South-Central SAS Users’ Group, MidWest SAS Users Group,
and the Pacific Northwest SAS Users Group. This paper provides an
overview of the project and the progress that has been made on this effort.
2:00 p.m.
GEN_ETA2: A SAS® Macro for Computing the Generalized
Eta-Squared Effect Size Associated with Analysis of
Variance Models
Patricia Rodriguez de Gil, University of South Florida
Thanh Pham, University of South Florida
Patrice Rasmussen, 5336 Clover Mist Drive
Jeanine Romano, University of South Florida
Yi-Hsin Chen, University of South Florida
Jeffrey Kromrey, University of South Florida
Paper 223-2013
Measures of effect size are recommended to communicate information on
the strength of relationships between variables. Such information
supplements the reject / fail-to-reject decision obtained in statistical
hypothesis testing. The choice of an effect size for ANOVA models can be
confusing because indices may differ depending on the research design as
well as the magnitude of the effect. Olejnik and Algina (2003) proposed the
generalized eta-squared effect size which is comparable across a wide
variety of research designs. This paper provides a SAS® macro for
computing the generalized eta-squared effect size associated with analysis
of variance models by utilizing data from PROC GLM ODS tables. The paper
provides the macro programming language, as well as results from an
executed example of the macro.
2:00 p.m.
2:00 p.m.
Anh Kellermann, USF
Aarti Bellara, University of South Florida
Diep Nguyen, University of South Florida
Eun Sook Kim, University of South Florida
Linking Laboratory Data To Submission Documents
Using SAS® Technologies
Dongmin Shen, Merck & Co, Inc
Paper 225-2013
Merck is a global pharmaceutical company and so the sources of our data
are global. Having the ability to link and transfer massive amounts of
analytical data from various data sources into submission documents in an
efficient and reproducible way is critical to producing successful regulatory
submissions. SAS® technologies have been used to create various solutions
ranging from data extractions, to data transformations, to documents
generated in support of simultaneous worldwide new drug applications.
2:00 p.m.
Linking Laboratory Data To Submission Documents
Using SAS® Technologies
Miu Ling Lau, Merck & Co.
Paper 225-2013
Merck is a global pharmaceutical company and so the sources of our data
are global. Having the ability to link and transfer massive amounts of
analytical data from various data sources into submission documents in an
efficient and reproducible way is critical to producing successful regulatory
submissions. SAS® technologies have been used to create various solutions
ranging from data extractions, to data transformations, to documents
generated in support of simultaneous worldwide new drug applications.
2:00 p.m.
Presenting Business Cases That Contain Complex,
Technical Information to a Varied Audience
Stephen Moore, US Census Bureau
Lori Guido, US Census Bureau
Paper 226-2013
The U.S. Census Bureau has a SAS® user base of approximately 2,600 users,
which the Software Application Branch (SADB) of the Applications Services
Division supports. In order to obtain the resources needed to provide the
support the users community needed, we had to figure out how to herd
cats. We had to gather information, enlist help from many sources, and get
everyone involved in the effort on the same level of understanding and
agreement. This paper describes the method SADB used to justify the
expansion of the Census SAS Support area from two to eight people. This
paper focuses on the following topics: the Census SAS support model, issue
definition, issue leveling, and communication strategy.
Variance Heterogeneity and Non-Normality: How the
SAS® TTEST Procedure Can Keep Us Honest
Paper 228-2013
The independent samples t-test is one of the most used tests for detecting
true mean differences. The SAS® System provides PROC TTEST, which is an
easy way to conduct a test for the difference between two population
means by assuming homogeneity of variance or avoiding it. However, the ttest and its alternatives (Satterthwaite's approximate test and conditional ttest) assume population normality; therefore, questions about the
performance of conditional testing when the assumption of normality is
not met remain. This paper describes previous research on preliminary tests
under the normality assumption, extends this research to the evaluation of
conditional testing to departures of normality, and provides guidance to
researchers on the proper use of this test with non-normal, heteroscedastic
population distributions.
2:00 p.m.
Here Comes Your File! File-Watcher Tool with
Automated SAS® Program Trigger
Rajbir Chadha, Cognizant Technology Solutions
Paper 229-2013
This paper talks about a file-watcher tool (UNIX Shell Script) that searches
for files and checks when they were last updated. Parameters to the filewatcher tool are supplied using a 'wrapper' script. Script is scheduled using
a 'CRON' scheduler in UNIX. Once file is found, SAS program is triggered. If
file is not found or not updated the script terminates. Tool sends out emails when files are available and when SAS program completes execution
or script terminates. In case of errors, users can refer to the file-watcher logs
at a location specified in the CRON file. The file-watcher tool reduces
average wait time and manual effort for users by automating most of the
process, allowing them to focus on other pressing tasks.
2:00 p.m.
Why the Bell Tolls 108 Times? Stepping through Time
with SAS®
Paper 230-2013
For many SAS® programmers, the use of SAS date and datetime variables is
often very confusing. This paper addresses the problems that the most of
programmers have. It starts by looking at the basic underlying difference
between the data representation and the visual representation of date,
datetime, and time variables. From there, it discusses how to change data
representations into visual representations through the use of SAS formats.
The paper also discusses date arithmetic first by demonstrating the use of
simple arithmetic to increment dates; then by moving on to SAS functions
which create, extract, and manipulate SAS date variables. This paper is
introductory and focuses on new SAS programmers; however, some
advanced topics are also covered.
19
2:00 p.m.
2:00 p.m.
Implementing CDISC, SDTM, and ADaM in a SAS®
Environment
A Preventive Approach for Automatic Checking of CDISC
ADaM Metadata to Detect Noncompliance
Pankaj Bhardwaj, Tata Consultancy Services
Paper 231-2013
Key challenges for regulatory bodies like FDA are non-standardized data
(almost 50% of the submissions) and its non-traceability. Reviewers cannot
streamline their review processes. A lot of work is happening in this
direction, and there is the expectation that all submissions will need to be
in standardized format by 2015 or so. This paper helps in building a
metadata-oriented, flexible, GUI-based solution for implementing the
CDISC and ADaM standards in a SAS® environment with following steps: 1.
Efficiently set up CDISC and ADaM metadata in SAS data sets, considering
important aspects like CDISC amendments and customization. 2. SAS
coding environment for handling legacy, ongoing and future trials. 3.
Generalized SAS validation codes for validation at the source data, SDTM,
and ADaMs level. 4. The submission deliverables.
2:00 p.m.
Weathering the Storm: Using Predictive Analytics to
Minimize Utility Outages
Mark Konya, Ameren Missouri
Kathy Ball, SAS
Paper 232-2013
Due to ever-increasing customer service expectations an ongoing
challenge for utilities is maintaining and improving the reliability of their
electric distribution systems. With significant numbers of transformers and
meters at risk of losing power during major storms, how can a Distribution
Engineer make sense of thousands of data points to prevent outages before
a storm occurs and, for customers whose power is interrupted during a
storm, restore service faster? Distribution Optimization equips utility
engineers and dispatchers to predict which assets will be affected by
storms while optimizing the placement of crews, thus decreasing outage
restoration times. Combining geospatial visualization with predictive
analytics, the predictive enterprise utility can shorten outages from weather
events and identify weak points in the electrical distribution system thus
preventing future outages.
2:00 p.m.
What Score Should Johnny Get? Missing_Items SAS®
Macro for Analyzing Missing Item Responses on
Summative Scales
Paper 233-2013
Missing data are usually not the focus of any given study but researchers
frequently encounter missing data when conducting empirical research.
Missing data for Likert-type response scales, whose items are often
combined to make summative scales, are particularly problematic because
of the nature of the constructs typically measured, such as attitudes and
opinions. This paper provides a SAS® macro, written in SAS/IML® and SAS/
STAT®, for imputation of missing item responses that allows estimation of
person-level means or sums across items in the scale. Imputations are
obtained using multiple imputation (MI), single regression substitution
(SRS), relative mean substitution (RMS), and person mean substitution
(PMS). In addition, the results of a simulation study comparing the accuracy
and precision of the imputation methods are summarized.
20
Min Chen, Vertex Pharmaceuticals, Inc.
Xiangchen Cui, Vertex Pharmaceuticals, Inc.
Tathabbai Pakalapati, CYTEL INC.
Paper 234-2013
The ADaM programming specification serves as the primary source for
ADaM programming, Define.xml, and reviewer guide. It should meet FDA
requirements and follow CDISC ADaM guidelines. OpenCDISC Validator is a
very useful tool to check the compliance with CDISC models. Sometimes it
is too late and/or costly to fix the errors identified by the tool. This paper
introduces a preventive approach to check metadata compliance with
ADaM guidelines at an earlier stage even before actual ADaM data set
programming thereby avoiding the waste of time and resources for
correction at a later stage. It also automatically ensures the consistency of
variable attributes between ADaM data sets and the define files, which
guarantees technical accuracy and operational efficiency.
2:00 p.m.
Building Traceability for End Points in Analysis Data Sets
Using SRCDOM, SRCVAR, and SRCSEQ Triplet
Paper 235-2013
To be compliant with ADaM Implementation Guide V1.0, traceability
features should be incorporated to possible extent in analysis data sets.
SRCDOM, SRCVAR, and SRCSEQ triplet are used to establish data point
traceability in ADaM data sets. This paper provides various examples of
applying the triplet to establish traceability in efficacy ADaM data sets, and
shows the art of applying the triplet to different scenarios.
2:00 p.m.
Building Traceability for End Points in Analysis Data Sets
Using SRCDOM, SRCVAR, and SRCSEQ Triplet
Qunming Dong, Vertex
Paper 235-2013
To be compliant with ADaM Implementation Guide V1.0, traceability
features should be incorporated to possible extent in analysis data sets.
SRCDOM, SRCVAR, and SRCSEQ triplet are used to establish data point
traceability in ADaM data sets. This paper provides various examples of
applying the triplet to establish traceability in efficacy ADaM data sets, and
shows the art of applying the triplet to different scenarios.
2:00 p.m.
SAS® Admin's Best Friend - The Set-up and Usage of
RTRACE Option
Airaha Chelvakkanthan Manickam, Cognizant Technology
Solutions
Srikanth Thota, Cognizant Technology Solutions
Paper 236-2013
The SAS® license of any organization includes various SAS components
such as SAS/STAT®, SAS/GRAPH®, SAS/OR®, etc. How does a SAS
Administrator know how many of the licensed components are actively
used, how many SAS users are actively utilizing the server, and how many
SAS data sets are frequently referenced? These questions help a SAS
administrator make important decisions such as controlling SAS licenses,
removing inactive SAS users, purging long-time non-referenced SAS data
sets, etc. SAS provides a system parameter called RTRACE to answer these
questions. The goal of this paper is to explain the set-up of the RTRACE
parameter and to explain its usage in making the SAS administrator's life
easy. This paper is based on SAS® 9.2 running on AIX 6.1 operating system.
2:00 p.m.
SAS® Stored Processes Logging
Bhargav Achanta, Reata Pharmacueticals
Paper 237-2013
You and your colleagues work very hard to create stored processes and
deliver them to various departments in your organization to review the
reports on regular basis and on time. Have you ever wondered how many
of the reports you provide to the audience are actually being used? This
paper presents a neat way to identify who ran the stored processes and
what time they have run the stored processes by scanning all the stored
process server log files and generates a list report and a frequency report.
2:00 p.m.
Creating a Management-Friendly HTML Report Using
SAS® ODS Markup, Style Sheets, and JavaScript
Rosely Flam Zalcman, Center for Addiction and Mental
Health
Robert Mann, 33 Russell St
Paper 238-2013
SAS® ODS output of 127 individual tables using SAS tagsets are integrated
into a single portable active HTML file. Hyperlinks and embedded
JavaScript menus provide easy access to both Client Satisfaction statistics
(15 pages) and Service Provider analysis (112 pages).
2:00 p.m.
Implementation of Slowly Changing Dimension to Data
Warehouse to Manage Marketing Campaigns in Banks
2:00 p.m.
Feature Extraction and Rating of a Smartphone
Photosharing Application Using SAS® Sentiment
Analysis Studio
Siddhartha Reddy Mandati, Oklahoma State University
Anil Kumar Pantangi, Oklahoma State University
Sahithi Ravuri, Oklahoma State University
Paper 241-2013
Smartphone users often have to read many online reviews to find out about
an application's feature. Online reviews usually provide an overall numeric
rating using the Likert or semantic scale, but these reviews do not fully
reveal the sentiments of customers. In this paper, the website Google Play is
considered. Google Play is a dedicated portal for all Android paid and free
applications. SAS® Sentiment Analysis Studio is used to predict a review as
either positive or non-positive. To extract features of the application, builtin manual rules in the rule-based model are used. In this data, the rulebased model outperformed the statistical and hybrid model. The best
model helps categorize each review of the application by its features and its
rating.
2:00 p.m.
V is for Venn Diagrams
Kriss Harris, SAS Specialists
Paper 243-2013
Would you like to produce Venn diagrams easily? This poster shows how
you can produce stunning two-, three-, and four-way Venn diagrams by
using the SAS® Graph Template Language, in particular the DRAWOVAL and
DRAWTEXT statements. From my experience, Venn diagrams have typically
been created in the pharmaceutical industry by using Microsoft Excel and
PowerPoint. Excel is used to first count the numbers in each group, and
PowerPoint is used to generate the two- or three-way Venn diagrams. The
four-way Venn diagram is largely unheard of. When someone is brave
enough to tackle it manually, then working out the numbers that should go
in each of the 16 groups and inputting the right number into the right
group is usually done nervously!
Lihui Wang, SIngapore Management University
Michelle Cheong, Singapore Management University
2:00 p.m.
In this paper, we illustrate the concept of the slowly changing dimension
and how it can be utilized in an innovative manner in the data warehouse
of a bank to update and maintain campaign records of customers.
Paper 245-2013
Paper 239-2013
SAS® Grid Job Submission and Monitoring from the SAS®
Information Delivery Portal
Adolfo Lopez, Valence Helath
As part of the implementation of SAS® Grid Computing at Valence Health,
we realized that users would need a simple and straightforward way to
submit SAS® programs to the grid from their desktops. While the SAS® Grid
Manager Client Utility provides this functionality. it requires that additional
software be installed on the client computer and that the user be
comfortable with a command line interface. To save time and effort, we
provided users with the ability to batch submit jobs to the grid and monitor
them via the SAS® Information Delivery Portal. This method provided the
functionality with minimal work and reduced the maintenance required to
ensure that the delivered solution met the needs of the majority of our
users.
21
2:00 p.m.
2:00 p.m.
Using Text Analysis to Gain Insight into Organizational
Change
Feature-Based Sentiment Analysis on Android App
Reviews Using SAS® Text Miner and SAS® Sentiment
Analysis Studio
Musthan Kader Ibrahim Meeran Mohideen, Oklahoma State
University
Jiawen Liu, Oklahoma State University
Gary Gaethe, U of Iowa
Douglas Van Daele, University of Iowa Healthcare
Paper 246-2013
Businesses often implement changes to improve customer satisfaction,
increase revenue, or improve profitability. The best situation occurs when a
business can measure the impact of the change before and after making
organizational changes. This research analyzes data from a survey of more
than 30,000 patients from a midwestern university teaching hospital. We
consider the impact of two very different changes: a move from free
parking to paid parking in 2009, and the implementation of a new online
portal designed so that patients can access their medical information. We
first analyzed the quantitative data using a key business metric and then
applied text mining and sentiment mining analysis procedures using the
qualitative data to gain deeper insights.
2:00 p.m.
So Many Films, So Little Time
Lisa Eckler, Lisa Eckler Consulting Inc.
Paper 247-2013
The Toronto International Film Festival ("TIFF") is an annual event, screening
a huge variety of new films for the international film industry as well as the
general public. The number of choices means selecting which films to order
tickets for can be overwhelming. I suffer the occupational hazard of
considering every logic problem in terms of SAS® code. Here we explore
how to use some very simple code to explore scheduling options which will
support decision-making with the goal of seeing the most films from a
priority list in the most enjoyable way. While many of us use SAS for
efficiency in our work, this is a small example of how it can also be
beneficial for personal time.
2:00 p.m.
SAS® ODS Graphics Designer - The Next Step in Amazing
Data Visualization
Christopher Battiston, Hospital For Sick Children
Admit it. You are swamped, overwhelmed, and desperate to find more
efficient ways of doing things. But who has the time to learn something
new? This poster won't be able to help for the majority of these issues. It
will help you become a more effective and efficient data visualization
expert, freeing up at least enough of your time to get a sandwich (and
maybe even eat it). SAS® ODS Graphics Designer is highlighted, showing
various examples with a generic step-by-step approach. Not as basic as
Graph-N-Go and not nearly as complex as SAS® Enterprise Guide®, ODS
Graphics Designer is a tool that appeals to both novice and expert users.
22
Jiawen Liu, Oklahoma State University
Mantosh Kumar Sarkar, Oklahoma State University
Paper 250-2013
Sentiment analysis is a popular technique for summarizing and analyzing
consumer textual reviews about products and services. There are two major
approaches for performing sentiment analysis—the statistical model-based
approach and the Natural Language Processing NLP-based approach. In
this paper, text mining is applied first to extract the features of Android
apps. Next, the NLP approach for writing rules is used. Reviews of two
recent apps are considered; a widget app from the Brain& Puzzle category
and a game app from the Personalization category. Six hundred textual
reviews are extracted for each app from the Google Play Android App Store.
Testing results show that for both apps, the carefully designed NLP rulebased model outperforms the default statistical model for predicting
sentiments and providing deeper insights.
2:00 p.m.
Analysis of Change in Sentiments towards Chick-fil-A
after Dan Cathy’s Statement about Same-Sex Marriage
Using SAS® Text Miner and SAS® Sentiment Analysis
Studio
Swati Grover, Student
Jeffin Jacob, Student, Oklahoma State University
Paper 251-2013
Social media analysis along with text analytics is playing a very important
role in keeping a tab on consumer sentiments. Tweets posted on Twitter
are one of the best ways to analyze customers’ sentiments following any
post-corporate event. Although there are a lot of tweets, only a fraction of
them are relevant to a specific business event. This paper demonstrates
application of SAS® Text Miner and SAS® Sentiment Analysis Studio to
perform text mining and sentiment analysis on tweets written about Chickfil-A before and after the company’s president’s statement supporting
traditional marriage. We find there is a huge increase in negative
sentiments immediately following the company president’s statement. We
also track and show that the change in sentiment persists through an
extended period of time.
2:00 p.m.
Analyzing Partially Confounded Factorial Conjoint
Choice Experiments Using SAS/IML®
Song Lin Ng, Universiti Tunku Abdul Rahman
Paper 252-2013
A 2^8 partially confounded factorial design with two replicate was applied
to CCE in this study. In this study, all the responses were assumed to be
independent and hence the multinomial logit model follows. The log
likelihood is nonlinear and hence the newton-raphson method is needed to
estimate the parameters. PROC IML was used to generated the NewtonRaphson procedures. The result showed that all main effects and some of
the first-order interaction effects were significant.
2:00 p.m.
2:00 p.m.
Analyzing Partially Confounded Factorial Conjoint
Choice Experiments Using SAS/IML®
MIXED_FIT: A SAS® Macro to Assess Model Fit and
Adequacy for Two-Level Linear Models
Chin Khian Yong, Universiti Tunku Abdul Rahman
Paper 252-2013
A 2^8 partially confounded factorial design with two replicate was applied
to CCE in this study. In this study, all the responses were assumed to be
independent and hence the multinomial logit model follows. The log
likelihood is nonlinear and hence the newton-raphson method is needed to
estimate the parameters. PROC IML was used to generated the NewtonRaphson procedures. The result showed that all main effects and some of
the first-order interaction effects were significant.
2:00 p.m.
SAS Training for STD Grantees
Robert Nelson, CDC
Molly Dowling, CDC
Delicia Carey, CDC
Paper 253-2013
The efficient and effective use of STD surveillance data for programmatic
decision-making is critical to state and local STD programs. SAS software is
well-suited to help realize this goal and is available from CDC at no-cost to
state and local STD grantees. The Division of STD Prevention at CDC has
developed a SAS training course for STD grantees, SASSI, to help meet this
need. To make the training useful to users at all levels of experience, each
module is designed to stand alone. Realistic data and real-world examples
are used to help ensure relevance to the target audience and state and
local STD program staff were engaged at all phases of development. For
more information, visit http://www.cdc.gov/std/.
2:00 p.m.
A Comparison of Model Building via RPM in SAS®
Enterprise Guide® versus SAS® Enterprise Miner™
Srikar Rayabaram, Oklahoma State University
Mihaela Ene, University of South Carolina
Whitney Smiley, University of South Carolina
Bethany Bell, University of South Carolina
Paper 255-2013
When estimating multilevel models, it is important for researchers to make
sure their models fit their data. However, examining model fit can be quite
cumbersome. We have developed the macro MIXED_FIT to help researchers
assess model fit in a simple yet comprehensive way. Specifically, this paper
provides a SAS® macro that incorporates changes in model fit statistics [that
is, -2 log likelihood (-2LL), AIC, and BIC] as well as changes in pseudo-R2. By
using data from PROC MIXED ODS tables, the macro produces a
comprehensive table of changes in model fit measures and allows SAS
users to examine model fit in both nested and non-nested models, both in
terms of statistical and practical significance without having to calculate
these values by hand.
2:00 p.m.
Optimize SAS/IML® Software Codes for Big Data
Simulation
Chao Huang, Oklahoma State University
Yu Fu, Oklahoma State University
Paper 256-2013
Nowadays, real-world data volume keeps growing. Simulation also creates
large data sets. To speed up the processing of a large data set, vectorization
is a very useful code optimization skill for many matrix languages such as R,
MATLAB, and SAS/IML®. In this paper, three simulation examples in SAS/IML
are used to discuss the implementation of the latest functions and
operators from SAS/IML for vector-wise operations. The result shows that
applying vectorization in SAS/IML significantly improves the computation
performance. SAS® ODS graphics procedures are used to visualize the
results.
Paper 254-2013
2:00 p.m.
Today, most large organizations use analytics for better decision making.
Even with the widespread availability of point-and-click interfaces for
advanced predicting modeling and analytics software such as SAS®
Enterprise Miner™, building good predictive models still requires analysts to
pre-process and manipulate data. The job of pre-processing, configuring,
and comparing requires a person with deep statistical or data modeling
knowledge which large businesses can afford but this is not the case with
SMEs. SAS® contends that using RPM makes it very easy for a person with
minimal training in the area of statistics or data modeling to quickly
develop a reasonable predictive model. In this paper, we test this
contention by a controlled experiment.
Top 10 Most Powerful Functions for PROC SQL
Chao Huang, Oklahoma State University
Paper 257-2013
PROC SQL is actually not a standard SAS® procedure but a distinctive
subsystem with all features from SQL (structured query language).
Equipped with it, SAS upgrades to a full-fledging relational database
management system. In addition, PROC SQL always provides alternative
ways to manage data, besides the traditional DATA step and procedures.
SAS also supplies some goodies, such as its functions, to further strengthen
SQL operation by PROC SQL.
23
2:00 p.m.
Data Set Compression Using COMPRESS=
Srinivas Reddy Busi Reddy, Oklahoma State University
Musthan Kader Ibrahim Meeran Mohideen, Oklahoma State
University
Paper 258-2013
Due to increased awareness about data mining, text mining, and big data
applications across all domains, the value of data has been realized and is
resulting in data sets with large number of variables and increased
observation size. Often it takes enormous time to process these data sets,
which can have an impact on delivery timelines. In order to handle these
constraints, think of making a large data set smaller by reducing the
number of observations, variables, or both, or by reducing the size of the
variables, without losing any of its information. In this paper, we see how a
SAS® data set can be compressed by using the COMPRESS= system option.
We also discuss some techniques to make this option more effective.
2:00 p.m.
An Integrated Approach to Codebook Generation Using
SAS®, HTML/CSS, and the .NET Framework
Helen Smith, RTI International
Mai Nguyen, RTI International
Elizabeth Eubanks, RTI International
Shane Trahan, RTI International
Paper 259-2013
For large surveys, creating comprehensive codebooks presents many
challenges. Without automation, this process becomes highly laborintensive and error-prone with data in the codebook quickly becoming
stale and failing to accurately represent underlying data sets. Another
significant challenge is that information/data for codebooks can come from
multiple sources. Such sources can include but not be limited to
questionnaire specifications, questionnaire design systems, and other
relational databases or SAS® data sets containing pertinent data. Our poster
presents an integrated approach for codebook generation using modern
tools and technologies, including SAS dictionary tables and SAS Integrated
Object Model (IOM) for data management, HTML/CSS for codebook
presentation, and the .NET framework for integrating and tying disparate
pieces together into one formatted codebook.
2:00 p.m.
Investigating the Impact of Amazon Kindle Fire HD 7” on
Amazon.com Consumers Using SAS® Text Miner and
SAS® Sentiment Analysis
Srihari Nagarajan, SAS Institute
Hari harasudhan Duraidhayalu, Kavi Associates
Paper 261-2013
This paper demonstrates the application of text mining techniques to
collect, group, and summarize positive and negative opinions of a product.
Unfortunately for popular products there are too many reviews, making it
difficult to read through all reviews and make an informed decision. For this
purpose, we developed a tool using ASP.NET to extract 1,674 customer
reviews for Kindle Fire HD 7” from Amazon.com. On the Microsoft Excel
data set thus generated, text mining can be performed to summarize
customer comments by grouping related reviews into clusters. The Text
24
Parsing, Filter, Topic, and Cluster nodes are used, and outputs from every
node are discussed. Sentiment analysis is performed on the data set to
develop a model for classifying positive and negative reviews.
2:00 p.m.
Practical Application of SAS® Capabilities for Pharma
Goals and Performance Review
Ramya Purushothaman, Cognizant Technology Solutions
Paper 262-2013
This paper discusses a Pharma application that uses SAS® to leverage
internal and purchased information such as Sales and Marketing data
including drug prescriptions, dollar and unit demand, target prescribers,
and key customer account profiles to set goals, measure sales performance,
and identify trends across geography levels. The capability of SAS to handle
huge volumes of data seamlessly provides an advantage over other
technologies. The reusability of SAS macros makes SAS solutions extensible
across various brands, sales teams, and geography levels for reporting. All
of these tasks are performed through familiar Base SAS® procedures,
functions, statements, and options. The paper explains how the business
need is addressed using SAS by accessing, cleansing, and transforming
information.
2:00 p.m.
Do People Still Miss Steve Jobs As the CEO of Apple Inc.?
A Text Mining Approach: Comparing SAS® and R
Pranav Karnavat, Shanti Communication School
Anurag Srivastava, Decision Quotient
Paper 263-2013
Marketers need information on views, expressions, need and expectation of
people from social media to capitalize upon and satisfy needs and
expectation of the consumers. Twitter is a powerful social media website.
Tweets posted can be analyzed to get insights about relationships and
patterns hidden inside the textual data. In this paper tweets were collected
about Steve Jobs prior to and post his sad demise to find if customers still
miss him as the CEO of Apple Inc. using text mining technique in SAS and R.
Get tweet macro is used to fetch data from twitter in SAS while twitteR
package is to fetch data from twitter in R. To analyze data, SAS Text Miner
was used in SAS while tm package in R.
2:00 p.m.
Build Prognostic Nomograms for Risk Assessment Using
SAS®
Dongsheng Yang, Cleveland Clinic
Paper 264-2013
Nomograms from multivariable logistic models or Cox proportionalhazards regression are a popular visual plot to display the predicted
probabilities of an event for decision support. In this paper, we show how
to build a prognostic nomograms after fitting a multivariable model,
including how to assign points for each predictor under different situations
such as main effect, interaction, piecewise linear effects. Furthermore, we
also show how to use a power tool, graphic template language to construct
a nomogram Finally, a SAS® macro was developed to generate a
nomogram.
2:00 p.m.
2:00 p.m.
SAS® Enterprise Guide®: Implementation Hints and
Techniques for Insuring Success with Traditional SAS
Programmers
Life's a Song! Mining Country Music Topics Using SAS®
Text Miner
Roger Muller, Data-To-Events.Com
Paper 265-2013
Deovrat Kakde, Kavi Associates
Saurabh Ghanekar, Kavi Associates
Neetha Sindhu, Kavi Associates
There are many configuration options available in SAS® Enterprise Guide®
for both the product itself and the included advanced editor. There are also
numerous software products from SAS® that may or may not be licensed at
your site and greatly affect your workflow. Workflow options while
developing the code are numerous and range from simple line-by-line
execution up to and including the running of an entire project flow or
process. Storage of SAS code under development also deserves careful
thought. All of these topics and more are addressed to enable users to have
a very thorough non-frustrating first-time experience with SAS Enterprise
Guide. The presentation is aimed at users who have experience coding and
running SAS programs.
Paper 268-2013
2:00 p.m.
2:00 p.m.
Repairable Systems—No Longer the Stepchild of
Reliability!!! Repairable System Reliability Modeling
Using PROC RELIABILITY in SAS/QC® 9.3
Predicting Application Review Rating with SAS® Text
Miner
Deovrat Kakde, Kavi Associates
Vijitha Kaduwela, Kavi Associates
Paper 266-2013
Most assets are repairable in nature. These assets include transportation
systems such as trucks and locomotives, oil and gas drilling equipment, and
heavy engineering equipment such as earthmoving equipment. When
assets break down, they are repaired rather than replaced. The
measurement and characterization of repairable system reliability requires
a different set of statistical techniques as compared to a system that cannot
be repaired. The RELIABILITY procedure in SAS/QC® 9.2 allowed modeling
of repairable system reliability using the nonparametric mean cumulative
function (MCF). In SAS/QC 9.3, PROC RELIABILITY offers a much-needed
functionality to model recurrent event data by fitting a nonhomogeneous
Poisson process (NHPP). This paper illustrates the use of nonparametric
MCF and parametric NHPP to model reliability of critical subsystems of a
repairable system.
2:00 p.m.
Getting an Overview of SAS® Data in Three Steps
Shirmeen Virji, Oklahoma State University
Miriam McGaugh, Oklahoma State University
Paper 267-2013
For SAS programmers, one of the most important steps before
manipulating the dataset for further analysis is to get an overview of it. In
order to get an idea of the dataset, normally three areas are looked into:
variable names, statistical description, and relationship of one dataset with
other datasets within a library. The macro program introduced in this paper
writes out the names of all the variables present in a file of a particular
library, gives descriptive statistics of all the variables that are classified as
numeric, and draws a diagram to show the relationships among the
datasets. All three steps are performed by running just one program.
Rich lyrics, often with a message, are a hallmark of American country music.
Typical song topics in American country music include family, marriage,
divorce, cheating, finding love, losing love, heartbreak, happiness, drinking,
children, men, women, honky tonk, religion, politics and love of country.
This paper demonstrates the use of SAS® Text Miner to identify topics in
American country music. The lyrics of Country Music Television's (CMT) top
20 songs for the last 25 years were analyzed. The prominent topics as
identified by SAS Text Miner were compared against the tags of last.fm to
develop a measure of accuracy. The results were also validated with native
English-speaking experts.
Zhangxi Lin, The Rawls College of Business Administration,
Texas Tech University
Tianxi Dong, Rawls College of Business Administration,
Jonghyun Kim, Texas Tech University
Paper 269-2013
With the proliferation of text-based data on the Internet, there is a need for
dealing with the information overload. The large number of online user
reviews might present an obstacle to developers who want to know users'
feedback and to potential customers who are interested in applications.
Here we employ text analysis provided in SAS® Text Miner to predict the
overall and feature-based ratings for online application reviews. We use
examples from the Android Market and Apple Store, the real world of
online application stores. The findings might aid in promoting the sales of
applications by better satisfying customer demands.
2:00 p.m.
Predict the Delay in Your Airline Before They Do!
Hari harasudhan Duraidhayalu, Kavi Associates
Rajesh Inbasekaran, Kavi Associates
Paper 271-2013
This paper demonstrates the application of predictive modeling techniques
to predict the time delay in several domestic flights across the United
States. Delay in domestic flights has been a common phenomenon in the
United States and it would definitely be useful if a predictive methodology
was employed. The data set for this purpose was prepared by gathering the
past two years of data from a flight stats website. The weather details of
these airports were also collected to understand if the weather details can
be used for the prediction. By using modeling techniques such as multiple
regression, neural networks, and so on, the delay in airlines can be
predicted by knowing the airline carrier, origin, and destination airport.
25
2:00 p.m.
2:00 p.m.
Calculating Subset-Weighted Analysis in PROC
SURVEYFREQ and PROC GENMOD
An Improved Data Visualization Approach for
Monitoring and Analyzing Business Performance Using
SAS/QC® Control Chart and SAS/GRAPH® Annotate
Techniques
Jessica Hale, University of Oklahoma
Paul Darden, OUHSC
David Thompson, OUHSC College of Public Health
Paper 272-2013
Stratum-specific weighted analysis is available in SAS® procedures such as
PROC SURVEYMEANS and PROC SURVEYLOGISTIC, which include the
DOMAIN statement. However, other procedures that can model correlated
outcomes, including PROC GENMOD, do not. This presentation
demonstrates a method of assigning individual weights to each record in a
data set to perform weighted subset analysis on a correlated outcome
without creating domain variables or transferring analysis to a separate
program.
2:00 p.m.
An Exploratory Graphical Method for Identifying
Associations in Sparse r x c Contingency Tables
Martin Lesser, Feinstein Institute for Medical ResearchBiostatistics
Meredith Akerman, Feinstein Institute for Medical Research
Biostatistics
Paper 273-2013
We investigate a graphical method, based on scree plots, for visualizing
“significant” departures between observed and expected cell frequencies in
RxC contingency tables, with a large number of rows and/or columns. This
method is based on Snedecor and Cochran’s (1989) proposal to identify the
cells with the largest values of (O-E)2/E, known as the contribution to chisquare. The scree plot shows the contributions plotted in descending order,
so that the user can detect which cells contribute the significant
departures, thus suggesting where the null hypothesis of independence
may have been violated. This method may be useful in large sparse RxC
tables. We used the following SAS procedures to develop a macro for
producing the scree plot: PROC FREQ (chisq, cellchi2, deviation, ODS
output), PROC SQL, and PROC GPLOT.
2:00 p.m.
SAS® Enterprise Guide®: What's in It for the Long-Term
Highly Experienced SAS® Programmer
Roger Muller, Data-To-Events.Com
Paper 274-2013
What are the benefits of SAS Enterprise Guide® as the developmental
platform for highly experienced SAS programmers who have been writing
code for a long time? This paper demonstrates a number of features that
are available in SAS Enterprise Guide for not only programming, but
viewing SAS data sets, creating multiple report outputs, improving code
storage, providing project organization and management, and more. The
techniques will emphasize the importance of the work flows in SAS
Enterprise Guide. All of these are in a state-of-the-art Microsoft Windows
environment with full copy, cut, and paste capabilities. This presentation
will focus more on the benefit to the programmer rather than on the
feature itself. The bottom line will be "What is in it for me?"
26
Sheng Ding, Fedex
Baojian Guo, Fedex
Paper 275-2013
Monitoring and analyzing business performance have been proved
difficult, especially in today’s intricate business environment. Customized
Control Chart using SAS® annotate facility, however, can provide a very
useful data technique to visualize complicated business information with
manageable data visualization results. This poster introduces an improved
technique for statistical process control visualization. Combined SAS/QC®
control chart with SAS/GRAPH® annotate technique, the improved control
chart can be used to customize highlight out-of-control signals and
potential root causes. Furthermore, the authors applied customized
annotate library to leverage business impact with different the potential
root causes.
2:00 p.m.
10 SAS® Skills for Grad Student Survival: A Grad Student
“How-To” Paper
Elisa Priest, UNT HSC SPH
Paper 276-2013
Grad students learn the basics of SAS® programming in class or on their
own. Real-world research projects are usually complex and may require a
variety of different SAS tools and techniques for data exploration and
analysis. This paper is a culmination of the SAS challenges I overcame and
the SAS skills that I learned outside of the classroom. These 10 SAS skills
helped me to survive graduate school and successfully write a complex
simulation analysis in SAS for my dissertation.
2:00 p.m.
Speed it Up: Using SAS® to Automate Initial Discovery
Practices
Mariya Karimova, AdvanceMed, an NCI Company
Christine John, AdvanceMed, an NCI Company
Paper 277-2013
Healthcare investigations frequently begin with a tip containing very little
provider information. This presentation attempts to use SAS® to automate
the initial discovery process, turning a name into a full overview of the
provider. Multiple data sources are combined, which oftentimes require
fuzzy matching to resolve conflicting identifiers. The program utilizes INFILE
URL and SAS text functions to obtain meaningful information from various
websites. It further utilizes SAS ODS and SAS/GRAPH® to create a single
standard PDF report; which provides a visualization of provider billing
patterns, summarizes their affiliations, and embeds hyperlinks to original
web-based resources. Additional topics that are discussed include: creating
a script for multiple users, paramaterization, utilizing system variables, and
SAS® 9.2 to SAS® 9.3 conversion.
2:00 p.m.
2:00 p.m.
Performance Predictability By Using Social Profile in
Online P2P Lending Market
A Flexible Method to Apply Multiple Imputation Using
SAS/IML® Studio
Zhangxi Lin, The Rawls College of Business Administration,
Siming Li, Southwestern university of finance and
economics
Harshal Darade, Texas Tech University
Paper 279-2013
We study the borrower-, loan-, and group-related determinants of
performance predictability in an online P2P lending market by
conceptualizing financial and social strength to predict borrower rate and
whether the loan would be timely paid. The results of our empirical study,
conducted using a database of 9,479 completed P2P transactions in
calendar year 2007, provide support for the proposed conceptual model in
this study. The results showed that combining financial files with social
indicators can enhance the performance predictability in the P2P lending
market. Although social strength attributes do affect the borrower rate and
status, their effects are very small in comparison to the financial strength
attributes.
2:00 p.m.
Dashing out a Quick Dashboard of Graphs in SAS®
Alan Elliott, UT Southwestern
Linda Hynan, University of Texas Southwestern Medical
Center
Paper 280-2013
In a world overwhelmed with data, a challenge of a data analyst confronted
with a new data set is to produce quick and concise initial comparisons that
provide information about data distributions as well as quick statistical
comparisons on primary factors of interest. This paper combines summary
analysis graphs that incorporate statistical results in a matrix/dashboard
format on a single, concise page. SAS® users familiar with basic SAS
programming techniques will be able to produce these dashboards of
graphic results.
2:00 p.m.
Adolescent Smoking and Development of Long-Term
Habits: A Longitudinal Analysis in SAS®
Elizabeth Leslie, Kennesaw State University
Paper 281-2013
This study was an investigation into the impact of early adolescent smoking
on adult smoking habits of National Longitudinal Survey of Youth 1997
Participants over the course of 13 years. The data was from a survey
consisting of 1,212 individuals interviewed once a year for 13 years (1997 to
2009) with the frequencies and amounts of cigarettes smoked recorded.
SAS® was used for the analysis and SAS arrays, do loops and macros were
used in structuring the data. There is significant evidence that smoking
habits increase over time, sex, and age when started smoking have an
effect on number of cigarettes smoked, and the number of cigarettes
increases as the number of peers who smoke and does drugs increases.
Xue Yao, University of Manitoba
Lisa Lix, University of Manitoba
Paper 283-2013
Multiple imputation has been widely used for dealing with missing data
and measurement error problems in various scientific fields. SAS/STAT®
software offers the MI and MIANALYZE procedures for creating and
analyzing of multiple imputation data. Imputation methods in PROC MI can
be used for either continuous or classification variable with the monotone
missingness pattern and only for continuous variable with the arbitrary
missingness pattern. This paper provides an imputation method using SAS/
IML® Studio for the arbitrary missingness pattern with classification
variable. Implementing this method expands the ability to conduct
multiple imputation using SAS®.
2:00 p.m.
Growth Spline Modeling
Matthew Schuelke, Air Force Research Laboratory
Robert Terry, University of Oklahoma
Eric Day, University of Oklahoma
Paper 285-2013
In this paper we will present an extensible, hybrid statistical approach
comprised of spline modeling and growth modeling which allows for an
examination of how the relative antecedent contributions to an outcome
change through time while simultaneously controlling for past effects.
2:00 p.m.
Gee! No, GTL! Visualizing Data With The SAS Graph
Template Language
Ted Conway, Self
Paper 286-2013
When you need to produce a grid of related graphs with minimum coding,
PROC SGPANEL is hard to beat. But eventually you'll run into a situation that
demands more precise control over the output. Perhaps there are unusual
scaling/formatting requirements. Or information needs to be presented in a
specific order. Or things need to be clarified via annotations or other
markup. That's where the Graph Template Language (GTL) can help. In this
paper, we'll see how GTL can be used to create a customized grid of time
series plots from segments and measures found in the TOTARRESTS sample
data set. This may be of interest to all skill levels. It requires Base SAS, SAS
GTL, and the SAS Macro Facility on UNIX or the PC.
2:00 p.m.
Utilizing SAS® for the Construction of Preassembled
Parallel, Computerized Fixed-Test Forms under Item
Response Theory Framework
Yi-Fang Wu, Iowa Testing Programs, University of Iowa
Paper 287-2013
The preassembled, parallel computerized fixed-test (CFT) forms are among
the most popular computer-based testing models. In item response theory,
test information function plays a dominant role for designing and
comparing measurement precision of CFT forms. The current paper
develops an automated procedure by utilizing SAS® software and
procedures (i.e. PROC IML, PROC SQL, SAS/GRAPH®, GTL, and ODS) for
27
constructing the CFT forms. The purpose is to demonstrate an efficient way
to obtain test and item information functions for the CFT forms and to plot
the test and item characteristic curves along with informative summary
statistics. Also, the paper investigates how measurement precision relates
to conventional item statistics. For test developers and practitioners, the
handy automated procedure through SAS and informative results are both
provided.
2:00 p.m.
2:00 p.m.
Paper 498-2013
A SAS® Macro Application for Efficient Interrupted Time
Series (ITS) Analysis Using Segmented Regression
Usefulness of text mining is now accepted worldwide to produce effective
knowledge and valuable insights of any business. Bing It On is an online
challenge offered by Microsoft allowing blind comparison of the search
results by Bing and Google. Microsoft claimed that users have chosen Bing
over Google nearly 2:1 times in these tests. Regarding this, there were
positive, negative, and mixed reactions from the vast user group, visible in
their tweets. In this research, we have collected relevant tweets using the
%GetTweet macro, and applied text mining to the data set using the SAS®
Text Miner® tab of SAS® Enterprise Miner™ 7.1 to summarize and portray the
general public opinion about this challenge and those two giant search
engines.
Sreedevi Thiyagarajan, Stanford University
Paper 288-2013
A comparison between a SAS® macro application and an existing software
tool (Joinpoint software) was conducted to identify the most efficient
software application to do a segmented regression for doing an interrupted
time series (ITS) analysis for asthma trends over time. The SAS macro
developed using the SAS 9.3 procedures NLIN and REG, when compared
with the Joinpoint software for an interrupted time series (ITS) analysis has
given an output similar to the latter and showed better running time,
efficiency as well as the time required to prepare the data sets, and total
analysis time.
2:00 p.m.
Using PROC FORMAT and Other Little Tweaks to Enable
PROC TABULATE’s Hidden Capabilities of Optimizing
Statistical Reporting
Heli Ghandehari, Baxter BioScience
Victor Lopez, Baxter Healthcare Corporation
Paper 289-2013
PROC TABULATE is arguably the most efficient approach for calculating
statistics and generating output, all within one procedure. However,
developers must often stray from PROC TABULATE when display
specifications require values to be reported as concatenated pairs. For
example, a common reporting requirement is for a mean and standard
deviation to be grouped within a single cell, with the latter enveloped by
brackets. Similarly, a range could be requested with the minimum and
maximum delimited by a dash, or perhaps a confidence interval nestled
within parentheses. The combinations are endless, but the underlying
solution is simple and universal. This paper demonstrates the utility of
PROC FORMAT’s PICTURE statement when applied in combination with
PROC TABULATE’s computational and reporting capabilities to create
customized statistical tables.
2:00 p.m.
Making it Happen: A Novel Way to Construct, Customize
and Implement Your SAS® Enterprise BI User
Enablement Framework
Tawney Moreno-Simon, Centers for Medicare & Medicaid
Services (CMS)
Vivek Seth, Computer Sciences Corporation - CSC
Paper 291-2013
Laying a solid foundation for user enablement is the holy grail of BI tool
implementation. Yet almost two-thirds (64%*) of BI Tool implementations
rate the success of user enablement initiatives “average” or lower. New BI
tool implementations struggle even further, with more than half (52%*)
rated as “fair” or “poor.”
28
Application of Text Mining in Tweets to Analyze General
Opinion about “Bing It On” Challenge by Microsoft
Shreya Sadhukhan, Oklahoma State University
Taufique Ansari, Oklahoma state university
2:00 p.m.
Impact of London Olympics According to Tweeters
Paper 499-2013
Overspending money on Olympics by the host country with the hope of
giving a huge boost to the economy is an age old phenomenon. The
purpose of this paper is to analyze the public sentiment on the economic
impact of London Olympics through tweets. SAS® Text Miner is employed
to summarize the collected tweets and classify them into different clusters.
Additionally, SAS® Sentiment Analysis Studio is used to corroborate our
findings and create a trend that tracks changes of public sentiments during
the London Olympics.
Quick Tips — Room 2003
2:00 p.m.
OUT= Is on the Way Out - Use ODS OUTPUT Instead
Stanley Fogleman, HARVARD CLINICAL RESEARCH
INSTITUTE
Paper 307-2013
There are, as a general rule, two methods to create a SAS data set from
procedural output. The more traditional one is the OUT= statement. This
feature is being replaced by the ODS OUTPUT statement as new capabilities
are added to procedures. In the future, only existing variables (generally in
SAS releases prior to SAS 9.1) will be available on the OUT= statement.
Therefore, it behooves the day-to-day SAS programmer to become familiar
with the new syntax.
2:15 p.m.
Implementing Metadata-driven Reports with SAS®
Stored Processes
Toby Hill, Charles Marcus Group Services
Paper 293-2013
As more organizations that use SAS® software are implementing the full
Business Intelligence reporting suite, many SAS programmers are
becoming familiar with developing SAS Stored Processes to deliver reports
for the business. Developers are often required to implement content
security in the reports or provide additional functionality for users with
specific roles. How can all this be done? One approach is to make use of the
SAS metadata. This paper demonstrates some techniques that you can
apply to your SAS code in order to make use of the SAS metadata. This will
allow you to implement security and role-based access in your SAS Stored
Process reports and minimize the amount of changes required as new users
access the platform.
function in the SAS® DATA step provides users with a simple and effective
approach to getting JSON information into SAS data sets. In this paper, two
examples of using this technique are provided.
3:15 p.m.
Back Up Your Sources During Development: A Stack of
Base SAS® Scripts
Hans Sempel, Belastingdienst (Dutch Tax and Customs
Administration)
Paper 297-2013
If you’re a Base SAS® programmer and if you ever lost your code due to
system crashes or overwriting your code, this might be the solution. The
presented code provides a means of backing up your code during
development, you can use it to save increments or you can use it for
versioning and you can restore the code you’re working on to an earlier
version.
2:30 p.m.
3:30 p.m.
Time Series Data: Anatomy of an ETL Project
Dealing with End-of-line Markers in Text Data Shared
Across Operating Systems
Leonard Polak, Wells Fargo Technology and Operations
Group
Paper 294-2013
It’s one thing to study SAS® tools and another to apply them to actual
situations. In this paper, we follow along as web data is copied and
transformed--and ultimately made available to users.
2:45 p.m.
CLISTS: Improve Efficiency of TSO Applications Using
Mainframe SAS®
Russell Hendel, Centers for Medicare and Medicaid Services
Paper 295-2013
Have you been spending a few hours every month submitting several
dozen SAS® jobs to mainframe systems using an IBM TSO environment with
the Interactive System Productive Facility (ISPF)? You know that within SAS,
SAS macros can efficiently manage repetitive tasks; but how do you
manage repetitive tasks with JCL, the TSO control language? CLIST is
precisely what you need: It enables you to automate repetitive tasks that
use JCL and SAS. CLIST is an easy language to learn, requiring no former
knowledge and using only a handful of basic commands. We present
illustrative CLIST code covering basic groups of CLIST commands. People
already familiar with JCL and SAS who write jobs using both of them will
benefit from this presentation.
3:00 p.m.
Efficient Extraction of JSON Information in SAS® Using
the SCANOVER Function
Kyong Jin Shim, Singapore Management University
Paper 296-2013
JSON or JavaScript Object Notation is a popular data interchange format
that provides a human readable format. It is language independent and can
be read easily in a variety of computer languages. With the rise of Twitter
and other types of unstructured data, there has been a move to incorporate
this data as a way of disseminating information. Twitter currently provides a
simple API for users to extract tweets using the JSON format. Although SAS
does not currently have a direct way of reading JSON, the SCANOVER
Haoyu Gu, University of Michigan
Paper 326-2013
Different operating systems use different end-of-line markers. When
sharing data across operating systems, caution must be taken. In this paper,
two examples are used to show how to read and write text data created in
Microsoft Windows from UNIX or Linux. In the examples, the use of option
TERMSTR and DLM=200Dx are discussed. The programs are run using both
SAS/CONNECT® and batch mode.
3:45 p.m.
Data Review Information: N-Levels or Cardinality Ratio
Ronald Fehd, retired
Paper 299-2013
This paper reviews the database concept: Cardinality Ratio. The SAS®
FREQUENCY procedure can produce an output data set with a list of the
values of a variable. The number of observations of that data set is called NLevels. The quotient of N-Levels divided by the number of observations of
the data is the variable's Cardinality Ratio. Its range is in (0-1]. The
Cardinality Ratio provides an important value during data review. Four
groups of values are examined.
4:00 p.m.
Healthcare Claims Processing with Base SAS® through
Denormalization of ANSI 837 Format
Victor Shigaev, CDC
Paper 300-2013
Sometimes dealing with healthcare claims can be messy. As a result of
HIPAA, all health insurance claims must be submitted to insurance payers
using the ANSI X12 837 messaging standard. This standard creates a
compact hierarchical file for quick transmission between trading partners
but because of the really complex nested structure of the data this standard
is not always easy to read in and be analyzed. The paper will give a brief
introduction to the X12 837 messaging standard, provide users a simple
way to divide raw claims data by claims through de-normalization, and a
way to use SAS® as a main tool to process and analyze the claims data.
29
4:00 p.m.
Healthcare Claims Processing with Base SAS® through
Denormalization of ANSI 837 Format
Roberto Valverde, NCHS
Paper 300-2013
Sometimes dealing with healthcare claims can be messy. As a result of
HIPAA, all health insurance claims must be submitted to insurance payers
using the ANSI X12 837 messaging standard. This standard creates a
compact hierarchical file for quick transmission between trading partners
but because of the really complex nested structure of the data this standard
is not always easy to read in and be analyzed. The paper will give a brief
introduction to the X12 837 messaging standard, provide users a simple
way to divide raw claims data by claims through de-normalization, and a
way to use SAS® as a main tool to process and analyze the claims data.
4:15 p.m.
Accessing SAS® Code via Visual Basic for Applications
Jennifer Davies, Z, Inc
Paper 306-2013
SAS® software has functionality that applications such as Microsoft Access
or Excel do not have and vice versa. However, in some situations, Microsoft
applications are preferred by the user over SAS for a multitude of reasons.
This paper will discuss how to integrate the use of Microsoft applications
with the functionality of SAS programs. This becomes very important when
SAS® Business Intelligence is not available. Depending on how SAS is
installed in the user’s organization, the programmer may have to access
SAS on the PC or a server version of the application. This paper will explain
the two methods used for calling SAS code from Visual Basic for
Applications (VBA) Code (v6.5).
4:30 p.m.
Something for Nothing? Adding Group Descriptive
Statistics Using PROC SQL Subqueries
Paper 302-2013
Can you actually get something for nothing? With PROC SQL’s subquery
and remerging features, yes, you can. When working with categorical
variables, often there is a need to add group descriptive statistics such as
group counts, minimum and maximum values for further by-group
processing. Instead of first creating the group count, minimum or
maximum values and then merging the summarized data set to the original
data set, why not take advantage of PROC SQL to complete two steps in
one? With PROC SQL’s subquery and summary functions by the group
variable, you can easily remerge the new group descriptive statistics back
to the original data set.
4:45 p.m.
FCMP -- Why?
Lisa Eckler, Lisa Eckler Consulting Inc.
Paper 298-2013
PROC FCMP allows a SAS® programmer the opportunity to create userdefined functions in SAS. Prior to the availability of FCMP in SAS 9, SAS
macros or linked routines were often used to achieve a similar –- but less
elegant -– effect. This paper examines the advantages of FCMP over the
earlier alternatives and why it is therefore so valuable to the programmer.
30
Reporting and Information Visualization — Room
2002
2:00 p.m.
Virginia's Best: How to Annotate County Names and
Values on a State Map
Anastasiya Osborne, Farm Service Agency
Paper 354-2013
This paper describes a work project to annotate a Virginia state map with
long county names and National Agricultural Statistics Service (NASS) data,
using enhanced color and techniques to minimize map crowding.
Displaying text and numeric data by county on a state map is different from
displaying state-level data on a U.S. map. Long county names rather than
two-letter state abbreviations require additional effort by a programmer to
create a readable map. A SAS® program with %ANNOMAC, %CENTROID,
PROC GPROJECT, PROC GMAP, and a 20-pattern color scheme was
developed to create maps that showcased in color Virginia's top
agricultural counties. This paper is for intermediate-level programmers.
2:30 p.m.
Data Merging and Visualization to Identify Associations
Between Environmental Factors and Disease Outbreaks
Neeta Shenvi, Emory University
Xin Zhang, Emory University
Azhar Nizam, Emory University
Paper 355-2013
This paper describes data merging and visualization techniques for
epidemiological and environmental surveillance data. The ultimate goal is
to learn about associations between specific environmental factors and
disease outbreaks. In such studies, environmental and clinical surveys often
occur on different timelines. We illustrate data merging with PROC SQL to
merge environmental and clinical data with chronological lags. We use the
Graph Template Language (GTL) to demonstrate data visualizations and
correlations that enabled us to identify potential associations between
cases of the disease and environmental variables, with a variety of possible
lags.
3:00 p.m.
Introducing and Producing Thunderstorm or Rain-drop
Scatter Plots Using the SAS/GRAPH® Annotate Facility
Charlie Liu, Allergan, Inc.
Paper 357-2013
A new type of plot, the thunderstorm (or rain-drop) scatter plot is
introduced. Such a plot allows for viewing data with two or more values on
the y-axis corresponding to one value on the x-axis for each of several
subjects in a population. The resulting plot looks like rain-drops, with each
rain-drop representing data for a single subject. When data for many
subjects is plotted, it resembles a thunderstorm (hence the name). A
thunderstorm or rain-drop scatter plot is a useful tool for data visualization
and outlier detection. Using examples from clinical research, this paper
shows how to create a thunderstorm or rain-drop scatter plot by using the
SAS/GRAPH® annotate facility.
3:30 p.m.
5:00 p.m.
Do SAS® users read books? Using SAS graphics to
enhance survey research
A Concise Display of Multiple Response Items
Paper 367-2013
Surveys often contain multiple response items, such as language where a
respondent may indicate that she speaks more than one language. In this
case, an indicator variable (1=Yes, 0=No) is often created for each language
category. This paper shows how a concise tabulation of the count and
percent of respondents with a “Yes” on one or more indicator variables may
be obtained using PROC TABULATE and a MULTILABEL format. A series of
indicator variables is used to create a binary variable and its base-10
equivalent, and a MULTILABEL format is created to properly aggregate
observations with a “Yes” on two or more indicator variables. The BAND
function may also be used to easily subset observations with “Yes”
responses on certain combinations of the indicator variables.
Barbara Okerson, WellPoint
In survey research, graphics play two important but distinctly different
roles. Visualization graphics enable analysts to view respondent segments,
trends and outliers that may not be readily obvious from a simple
examination of the data. Presentation graphics are designed to quickly
illustrate key points or conclusions to a defined audience from the analysis
of the survey responses. SAS provides the tools for both these graphics
roles through SAS/Graph and ODS graphics procedures. Using a survey of
the Virginia SAS Users Group (VASUG) as the data source, this paper
answers the above question and more while illustrating several SAS
techniques for survey response visualization and presentation. The
techniques presented here include correspondence analysis, spatial
analysis, heat maps and others.
4:00 p.m.
An Innovative Approach to Integrating SAS® Macros
with GIS Software Products to Produce County-Level
Accuracy Assessments.
Audra Zakzeski, USDA NASS
Robert Seffrin, US Dept. of Agriculture
Paper 358-2013
The National Agricultural Statistics Service (NASS) produces an annual
geospatial informational data set called the Cropland Data Layer over the
U.S. detailing the land cover over each state while focusing on the vast
array of crops grown during the months of April through October. While
calculating an accuracy assessment of the land cover over an entire state is
a relatively simple process, calculating an accuracy assessment down to a
county- or crop-specific level can be extremely time-consuming. To simplify
the process, NASS created an innovative SAS® program integrating the
efficiency of the SAS Macro language with the geospatial analytical
capabilities of the GIS program ERDAS Imagine. The procedure is operated
using a SAS/AF® platform allowing analysts to easily investigate countylevel information.
4:30 p.m.
How to Become a GTL/PROC Template Jedi
Paper 359-2013
This tongue-in-cheek paper will bring together Star Wars and SAS®,
answering (at least potentially) how would SAS have been used a long time
ago in a galaxy far, far away? Using PROC TEMPLATE, GTL, and ODS,
examples will be shown of reports that could have been used by the Empire
and the Rebel Alliance. Topics will include creating reports for mobile
devices, bringing in images into the reports, and creating dynamic reports without using Jedi mind tricks on anyone!
Patrick Thornton, SRI International
Paper 360-2013
5:30 p.m.
SAS Metadata Reporting: Extracting Invaluable
Information from SAS® Metadata
Jugdish Mistry, J2L Limited
Paper 518-2013
We can see a trend in the past few releases of SAS® software; there is a big
emphasis on using and moving to using more and more metadata. It is the
one-stop place now, for all SAS applications, configuration, SAS® Data
Integration Studio, SAS® Business Intelligence, and GRID developments.
This wonderful method of storing data and managing SAS has no nice GUI
for getting this information out. So if we wanted a user list, the name of the
last person to update a SAS Data Integration flow, or list new jobs created in
the past week, we have to use the appropriate GUI and manually get this
information. This paper discusses how using SAS one could extract and
generate useful reports from metadata.
SAS Futures — Room 2018
2:00 p.m.
Your GPS for SAS® on the Cloud
Saravana Chandran, SAS
Paper 509-2013
“Cloud computing” is both mystifying and perplexing. Have you ever
wondered what cloud computing means to SAS® products and solutions?
This paper explores various possibilities of SAS products and solutions on
the cloud – public, private and hybrid – and walks through three case study
deployment models. Find out about the cloud lifecycle aspects of SAS
solutions and products.
2:30 p.m.
SAS® Virtual Applications in Your Cloud Infrastructure
Peter Villiers, SAS
Paper 510-2013
Many organizations, regardless of size, are investing in cloud
infrastructures. These infrastructures can be in-house private clouds,
commercially provided public clouds or a hybrid cloud consisting of both
types. Making purchased software run in the cloud takes time and effort to
get it right and includes its own set of challenges. If software companies
provided products as prebuilt, cloud-enabled applications, this would
benefit organizations choosing to deploy them. This paper describes an
31
approach to developing prepackaged (virtual) applications for the cloud. It
outlines how these virtual applications can be deployed into various cloud
providers and integrated with existing IT resources. It also shows how these
applications can be distributed globally to provide faster communication
with users, while still being managed centrally.
3:00 p.m.
You Like What? Creating a Recommender System with
SAS®
Wayne Thompson, SAS
Paper 511-2013
Recommendation engines provide the ability to make automatic
predictions (filtering) about the interests of a user by collecting preference
information from many users (collaborating). SAS® provides a number of
techniques and algorithms for creating a recommendation system, ranging
from basic distance measures to matrix factorization and collaborative
filtering. The “wisdom of crowds” suggests that communities make better
decisions than a handful of individuals, and as a community grows, the
better its decisions are. With enough data on individual community
participation, we can make predictions about what an individual will like in
the future based on what their likes and dislikes have been in the past.
4:00 p.m.
Statistics and Data Analysis — Room 3016
10:30 a.m.
What Is Business Analytics?
J. Michael Hardin, University of Alabama
Analytics has become the hot, “sexy” job of the new century. The demand
for individuals with skills and expertise in the area are in great demand,
with feature articles appearing in publications ranging from The New York
Times to The Harvard Business Review to The Wall Street Journal. However,
the area has not always been so well received, especially within some
academic areas. And, even today there still remains confusion and
disagreements over the implementation and interpretation of results
obtained from the “analytic process.” This presentation will examine the
history, development, and particularly the philosophy underlying the
analytic process. Insights will be provided as to theories to understanding
and interpreting the analysis process and the associated results.
2:00 p.m.
Making the Most of Your SAS® Investment in the Era of
Big Data
Being Continuously Discrete (or Discretely Continuous):
Understanding Models with Continuous and Discrete
Predictors and Testing Associated Hypotheses
Paper 512-2013
If you follow the headlines, it doesn’t take long to recognize that big data is
a big deal. And if the amount of data that you’re faced with exceeds your
organization’s capacity for accurate and timely decision making, then it’s a
big deal for you, too! By taking advantage of massively parallel compute
environments and in-memory processing, SAS® can help you expand the
boundaries of what’s possible and transform the way you do business. This
paper helps you understand how to leverage SAS High-Performance
Analytics to explore, visualize and analyze large volumes of data.
Often a general (or generalized) linear model has both discrete predictors
(included in the CLASS statement) and continuous predictors. Binary
variables can be treated either as continuous or discrete; the resulting
models are equivalent but the interpretation of parameters differs. In many
cases, interactions between discrete and continuous variables are of
interest. This paper provides practical suggestions for building and
interpreting models with both continuous and discrete predictors. It
includes some examples of the use of the STORE statement and PROC PLM
to understand models and test hypotheses without repeating the
estimation step.
Tonya Balan, SAS
Justin Choy, SAS
5:00 p.m.
MapReduce Anywhere with DS2
Doug Sedlak, SAS
Robert Ray, SAS
Gordon Keener, SAS
Cindy Wang, SAS
Paper 398-2013
With the rise in all things Hadoop and the MapReduce programming
paradigm, you might ask, “Is the SAS® programmer left behind?” No, the
parallel syntax of the DS2 language coupled with our in-database SAS
Embedded Process and pass-through technology will allow the traditional
SAS developer to create portable MapReduce-type algorithms that are
implicitly executed in massively parallel environments, including Hadoop.
The DS2 portable syntax allows parallel algorithms to be verified on the SAS
client using sample data before releasing them on your largest problems.
32
David Pasta, ICON Late Phase & Outcomes Research
2:00 p.m.
Using the QUANTLIFE Procedure for Survival Analysis
Guixian Lin, SAS
Bob Rodriguez, SAS
Paper 421-2013
The QUANTLIFE procedure implements quantile regression, which provides
a direct and flexible approach to modeling survival times without the
proportionality hazard constraint of the Cox model. In clinical studies,
quantile regression is helpful for identifying and distinguishing important
prognostic factors for patient subpopulations that are characterized by
short or long survival times. This paper compares the quantile regression
model with the Cox and accelerated failure time models, which are
commonly used in survival analysis. An understanding of the differences
between these approaches is essential for deciding which model to use in
practice. An example illustrates how to estimate regression parameters and
survival functions.
3:00 p.m.
change in price using six-month transaction-level data. Limitations and
prospects of the methods used are discussed. The inclusion of promotions
and prices of other products as covariates provides a better understanding
of the dynamics of price-demand relationships.
Computing Direct and Indirect Standardized Rates and
Risks with the STDRATE Procedure
4:00 p.m.
Yang Yuan, SAS
Paper 423-2013
In epidemiological and health care studies, a common goal is to establish
relationships between various factors and event outcomes. But outcome
measures such as rates or risks can be biased by confounding. You can
control for confounding by dividing the population into homogeneous
strata and estimating rate or risk based on a weighted average of stratumspecific rate or risk estimates. This paper reviews the concepts of
standardized rate and risk and introduces the STDRATE procedure, which is
new in SAS/STAT® 12.1. PROC STDRATE computes directly standardized
rates and risks by using Mantel-Haenszel estimates, and it computes
indirectly standardized rates and risks by using standardized morbidity/
mortality ratios (SMR). PROC STDRATE also provides stratum-specific
summary statistics, such as rate and risk estimates and confidence limits.
Estimating Harrell's Optimism on Predictive Indices
Using Bootstrap Samples
Yinghui Miao, NCIRE
Irena Cenzer, UCSF
Katharine Kirby, UCSF
John Boscardin, UCSF
Paper 504-2013
In aging research, it is important to develop and validate accurate
prognostic models whose predictive accuracy will not degrade when
applied in external data sources. While the most common method of
validation is split sample, alternative methods such as cross-validation and
bootstrapping have some significant advantages. The macro that we
present calculates Harrell's optimism for logistic and Cox regression models
based on either the c-statistic (for logistic) or Harrell's c (for Cox). It allows
for both stepwise and best subset variable selection methods, and for both
traditional and .632 bootstrapping methods.
3:00 p.m.
Good as New or Bad as Old? Analyzing Recurring
Failures with the RELIABILITY Procedure
Paper 424-2013
Creating and Customizing the Kaplan-Meier Survival
Plot in PROC LIFETEST
Bobby Gutierrez, SAS
You can encounter repeated failure events in settings ranging from repairs
of equipment under warranty to treatment of recurrent heart attacks. When
a system fails repeatedly, the risk of failure can change with each
subsequent failure—a unit once repaired or a patient once treated might
not be “good as new.” Analysis with the RELIABILITY procedure focuses
primarily on the mean cumulative function (MCF), which represents either
the average number of failures per unit over time or some related cost
measure. This paper describes how you can use PROC RELIABILITY to
estimate and compare MCFs. You can choose either a nonparametric or a
parametric approach. Features added to the procedure in SAS/QC® 12.1 are
highlighted.
3:30 p.m.
Price and Cross-Price Elasticity Estimation Using SAS®
4:00 p.m.
Warren Kuhfeld, SAS
Ying So, SAS
Paper 427-2013
If you are a medical, pharmaceutical, or life sciences researcher, you have
probably analyzed time-to-event data (survival data). One of several
survival analysis procedures that SAS/STAT® provides, the LIFETEST
procedure computes Kaplan-Meier estimates of the survivor functions and
compares survival curves between groups of patients. You can use the
Kaplan-Meier plot to display the number of subjects at risk, confidence
limits, equal-precision bands, Hall-Wellner bands, and homogeneity test pvalue. You can control the contents of the survival plot by specifying
procedure options with PROC LIFETEST. When the procedure options are
insufficient, you can modify the graph templates with SAS macros. This
paper provides examples of survival plot modifications using procedure
options, graph template modifications using macros, and style template
modifications.
Dawit Mulugeta, Cardinal Health
Jason Greenfield, Cardinal Health
Tison Bolen, Cardinal Health
Lisa Conley, Cardinal Health
Paper 425-2013
The relationship between price and demand (quantity) has been the
subject of extensive studies across many product categories, regions, and
stores. Elasticity estimates have also been used to improve pricing
strategies and price optimization efforts, promotions, product offers, and
various marketing programs. This presentation demonstrates how to
compute item-level price and cross-price elasticity values for two products
with and without promotions. We used the midpoint formula, the OLS
linear model, and the log-log model to measure demand response to
33
4:30 p.m.
5:00 p.m.
Are You Discrete? Patients' Treatment Preferences and
the Discrete Choice Experiment
Assessing Model Adequacy in Proportional Hazards
Regression
Beeya Na, ICON Late Phase & Outcomes Research
Eric Elkin, ICON
Paper 429-2013
The discrete choice experiment (DCE) was designed for use in economics
and marketing research to study consumer preferences. DCE has been
increasingly used in health care research as a method to elicit patient
preferences for characteristics of different types of treatments. In a DCE,
attributes with varying levels are defined for treatments. Respondents are
presented with pairs of hypothetical treatments that have different
combinations of attribute levels and are asked to choose their preferred
treatment. Analyzing the responses allows evaluation of the relative
importance of the attributes and the trade-offs that respondents are willing
to make between the attributes. This paper explains how to set up the data
and discusses how to use the PHREG and LOGISTIC procedures to
appropriately analyze the conditional logit model.
4:30 p.m.
Cox Proportional Hazard Model Evaluation in One Shot
Polina Kukhareva, University of North Carolina at Chapel Hill
Paper 428-2013
Cox proportional hazard models are often used to analyze survival data in
clinical research. This article describes a macro that makes producing the
correct diagnostics for Cox proportional hazard models fast and easy. The
macro has three advantages over performing all the diagnostics one by
one. First, it makes it easy to run diagnostics for a long list of similar models.
Second, it allows the specification of the variables for which diagnostics
should be run. Third, it produces a comprehensive list of plots and tables
necessary for evaluation of the Cox proportional hazard model assumptions
as recommended in the SAS® course “Survival Analysis Using the
Proportional Hazards Model.” This macro can help save hours of codewriting time for a programmer who performs survival analysis.
5:00 p.m.
Chi-Square and t-Tests Using SAS®: Performance and
Interpretation
Jennifer Waller, Georgia Health Sciences University
Maribeth Johnson, Georgia Health Sciences University
Data analysis begins with data cleanup, calculation of descriptive statistics,
and the examination of variable distributions. Before more rigorous
statistical analysis begins, many statisticians perform basic inferential
statistical tests such as chi-square and t tests to assess unadjusted
associations. These tests help guide the direction of the more rigorous
analysis. This paper uses example data to show how to perform chi-square
and t tests, how to interpret the output, where to look for the association or
difference based on the hypothesis being tested, and which next steps can
be proposed for further analysis.
34
Michael G. Wilson, Biostatistical Communications, Inc.
Proportional hazards regression has become an exceedingly popular
procedure for conducting analysis on right-censored, time-to-event data. A
powerful, numerically stable, easily generalizable model can result from
careful development of the candidate model, assessment of model
adequacy, and final validation. Model adequacy focuses on overall fitness,
validity of the linearity assumption, inclusion (or exclusion) of a correct (or
an incorrect) covariate, and identification of outlier and highly influential
observations. Due to the presence of censored data and the use of the
partial maximum likelihood function, diagnostics to assess these elements
in proportional hazards regression compared to most modeling exercises
can be slightly more complicated. In this paper, graphical and analytical
methods using a rich supply of distinctive residuals to address these model
adequacy challenges are compared.
Systems Architecture and Administration — Room
2006
10:30 a.m.
Virtualized Environment for SAS® High-Performance
Products
Tom Keefer, SAS
Rich Pletcher, SAS
Daniel Zuniga, SAS
Paper 459-2013
Virtualization technology has reached maturity, and many companies are
moving quickly to adopt these environments to support their entire IT
infrastructures. Customers are increasingly asking for reference
architectures and best practices for deploying the latest SAS® products in
virtual environments. SAS has built and tested a reference architecture that
supports a completely virtualized environment for SAS 9.3, SAS Grid
Manager and newer products such as SAS High-Performance Analytics and
SAS Visual Analytics. This paper shares results from performance testing
and best practices on how to plan, manage and deploy a successful
enterprise-class, virtualized SAS environment.
11:30 a.m.
Building a SAS® Grid Support Capability in the
Enterprise
Andy Birds, The Co-operative Banking Group
Chris Rigden, SAS
Paper 460-2013
Building SAS® Grid support capability within an organization’s IT support
function requires IT managers to consider many different aspects. There is a
need for SAS support to fit seamlessly into the Enterprise IT support model
and comply with IT policies based on standard frameworks such as ITIL,
while maintaining a level of business engagement that is far beyond that
which is required of traditional IT support teams. We will outline a practical
framework including the policies and procedures required to support a SAS
Grid in a way that provides a solid foundation that meets the immediate
and ongoing business requirements. We will discuss how to embed SAS
into an organization’s standard IT processes and how to ensure that active
business engagement is a standard activity.
3:00 p.m.
12:00 p.m.
Amy Peters, SAS
Bob Bonham, SAS
Zhiyong Li, SAS
Tips and Techniques for Deploying SAS® in an
Application Virtualization Environment
Chuck Hunley, SAS
Michael King, SAS
Casey Thompson, SAS
Rob Hamm, SAS
Paper 461-2013
Application virtualization is increasingly used by many organizations to
more easily deploy, maintain and manage their desktop applications. There
are many vendors and products on the market to choose from, including
VMWare, Citrix and Microsoft. Each vendor’s technology comes with its
unique set of features and nuances. What do you need to know to get SAS®
up and running? This paper explores some best practices, gotchas and
guidelines that will help you succeed when deploying and using SAS in an
application virtualization environment.
2:00 p.m.
Pi in the Sky: Building a Private SAS® Cloud
Andrew Macfarlane, SAS
Frank Schneider, Allianz Managed Operations and Services
SE
Paper 494-2013
In today's climate, cloud computing is a de facto term used in IT and cloud
capability a mandatory requirement for all software vendors. Based on reallife experience, this paper will discuss challenges, opportunities, and
options for developing and implementing a private SAS® cloud using SAS®
9.3. For the purposes of this paper, we focus on some essential concepts of
cloud computing including Multi Tenancy (Resource pooling); Scalability
and rapid elasticity of resources, Shared Services and the building blocks of
Platform as a Service, and suggest approaches for applying these concepts
within the SAS platform.
2:30 p.m.
Writing a Useful Groovy Program When All You Know
about Groovy Is How to Spell It
Jack Hamilton, Kaiser Foundation Hospitals
Paper 493-2013
SAS® is a powerful programming system, but it can't do everything.
Sometimes you have to go beyond what SAS provides. There are several
built-in mechanisms for doing this, and one of the newest is PROC GROOVY.
It sounds like a product of San Francisco's Haight-Ashbury, but it's actually a
programming language based on another product with San Francisco Bay
area roots, Java. You can think of it as a simplified, easier to use version of
Java -- simplified enough that you can put together a useful PROC GROOVY
program from Internet examples without knowing anything about the
language. This presentation focuses on handling directories and ZIP files,
but many other things are possible.
Monitoring 101: New Features in SAS® 9.4 for
Monitoring Your SAS® Intelligence Platform
Paper 463-2013
Ever needed an alert on SASWORK storage usage at 80 percent? Or known
when a SAS® user account has been locked out due to failed login
attempts? Or to understand the memory and swap usage of a computer
hosting the SAS Stored Process Server? What if you could see a complete
listing of process resource consumption for all physical machines hosting a
given SAS deployment? Learn about options for answering these questions,
including new tools in SAS 9.4 that autodiscover software resources
(including SAS servers and Web application servers) of your platform for
SAS Business Analytics. Discover how to use agents to collect metrics that
reflect availability, performance, utilization and throughput, giving you a
more proactive understanding of the operational state of your SAS
deployment.
3:30 p.m.
SAS® 9.3 Administration Guilty Pleasures: A Few of My
Favorite Things
Diane Hatcher, SAS
Paper 464-2013
Over the evolution of SAS® 9.3, SAS has continued to enhance and augment
its administration capabilities. Most of these capabilities are well-known
and welcome additions for SAS administrators, but there are some hidden
jewels that you may not be aware of. This paper reveals some “guilty
pleasures,” features that make the SAS environment easier to manage and
more robust, including stored process report, metadata-bound libraries,
authdomain option for libname, and special tricks with metadata folders.
4:00 p.m.
Knowing Me, Knowing My UI (ah-haa): Understanding
SAS® 9.3 and SAS® 9.4 Desktop and Web Client
Application Usage Within Your Organization
Simon Williams, SAS
Paper 465-2013
Understanding which groups of users are using which sets of desktop and
Web applications can help your organization increase efficiency while
reducing the risk to your operations. This paper details the importance of
understanding users and desktop application interactions, and the
methods by which these interactions can be defined and reported on.
Working examples using SAS® desktop and Web applications such as SAS®
Enterprise Guide®, SAS Data Integration Studio, SAS Management Console
and SAS Web Report Studio are presented.
35
4:30 p.m.
SAS Visual Analytics (SASVA) and SAS High-Performance
Analytics Server (SASHPAS) - Network Considerations
and Data Management/Governance
Nicholson (Nick) Warman, Hewlett-Packard (onsite @ SAS)
SAS Visual Analytics (SASVA) and SAS High-Performance Analytics Server
(SASHPAS) are an entirely new approach to information analysis and
management. With these products come data network challenges/issues
and data provisioning/strategy issues. This paper begins that focused
dialogue based on over 20 years of experience with SAS and as the
engineer responsible for the world-wide configurations used by the HP(TM)
sales force, one of only two companies authorized to sell hardware to
support SASVA/SASHPAS.
5:30 p.m.
SAS® Release Management and Version Control
John Heaton, Heritage Bank
Paper 467-2013
Release Management or Application Lifecycle Management is the process
of versioning and migrating code from one environment to another in a
controlled, auditable, and repeatable process. This paper looks at the
capabilities of the current SAS® 9.3 toolset to build an effective Release
Management process within your organization.
36
Applications Development — Room 2014
8:00 a.m.
Tips and Techniques for Moving SAS® Data to JMP®
Graph Builder for iPad®
Michael Hecht, SAS
Paper 008-2013
You have an iPad®. You have the JMP® Graph Builder app. But how do you
get your SAS® data set to it so you can use all the cool features that JMP
Graph Builder has to offer? This paper describes how to use the new JMP
engine in SAS to create a workflow that makes it easy to move your data
sets to the iPad.
9:00 a.m.
MACUMBA: Modern SAS® GUI Debugging Made Easy
Michael Weiss, Bayer Pharma AG
Paper 009-2013
MACUMBA is an in-house-developed application for SAS® programming. It
combines interactive development features of PC-SAS, the possibility of a
client-server environment and unique state-of-the-art features that were
always missing. This presentation covers some of the unique features that
are related to SAS code debugging. At the beginning, special code
execution modes are discussed. Afterwards, an overview of the graphical
implementation of the single-step debugger for SAS macros and DATA step
is provided. Additionally, the main pitfalls of development are discussed.
9:30 a.m.
Give the Power of SAS® to Excel Users Without Making
Them Write SAS Code
William Benjamin, Owl Computer Consultancy LLC
Paper 010-2013
Merging the ability to use SAS® and Microsoft Excel can be challenging.
However, with the advent of SAS® Enterprise Guide®, SAS® Integration
Technologies, SAS® BI Server software, JMP® software, and SAS® Add-In for
Microsoft Office; this process is less cumbersome. Excel has the advantages
of being cheap, available, easy to learn, and flexible. On the surface, SAS
and Excel seem widely separated without these additional SAS products.
But wait, BOTH SAS AND EXCEL CAN INTERFACE WITH THE OPERATING
SYSTEM. SAS can run Excel using the command and Excel can run SAS as an
“APPLICATION.” This is NOT DDE; each system works independently of the
other. This paper gives an example of Excel controlling a SAS process and
returning data to Excel.
10:00 a.m.
Automated Testing of Your SAS® Code and Collation of
Results (Using Hash Tables)
Andrew Ratcliffe, RTSL.eu
Paper 011-2013
Testing is an undeniably important part of the development process, but its
multiple phases and approaches can be under-valued. I describe some of
the principles I apply to the testing phases of my projects and then show
some useful macros that I have developed to aid the re-use of tests and to
collate their results automatically. Tests should be used time and again for
regression testing. The collation of the results hinges on the use of hash
tables, and the paper gives detail on the coding techniques employed. The
small macro suite can be used for testing of SAS® code written in a variety
of tools including SAS® Enterprise Guide®, SAS® Data Integration Studio,
and the traditional SAS Display Manager Environment.
10:30 a.m.
A Metadata-Driven Programming Technique Using SAS®
Xiyun (Cheryl) Wang, Statistics Canada
Paper 012-2013
In a typical SAS® system, validations on user inputs and setting defaults for
missing values in inputs are essential to ensure that the system can
continue its processing without errors. This paper describes, in a SAS
system, how to define validation rules as metadata for various type of
inputs such as library, data sets, etc., and how to register default values for
missing values for inputs as metadata. Furthermore, it illustrates how to use
those metadata to automate the validation processes and data imputation
process. It provides a SAS programming technique to ease system
development with efficiency, re-usability, easy maintainability, and coding
consistency.
11:00 a.m.
Knowing When to Start, Where You Are, and How Far
You Need to Go: Customized Software Tracks Project
Workflow, Deliverables, and Communication
Eric Vandervort, Rho
Paper 013-2013
In a clinical trials environment, projects can have multiple statisticians and
statistical programmers working on tables, listings, and figures, or
"displays", for project deliverables. Communication between the various
team members regarding when to program, validate, review, or revise
these displays is vital to the success of a project. This paper describes a
custom web-based application that stores relevant data about displays,
tracks programming and reviewing workflow, and provides a tool for
project-level management overview.
11:30 a.m.
Extension Node to the Rescue of the Curse of
Dimensionality via Weight of Evidence (WOE) Recoding
Satish Garla, SAS
Andrew Cathie, SAS
Paper 014-2013
Predictive models in data mining applications often involve very large data
sets that contain numerous categorical variables with large numbers of
levels. These models often suffer from the curse of dimensionality.
Enhanced weight of evidence (WOE) methods can be used to effectively
incorporate high-dimensional categorical inputs into a data mining model.
Weight of evidence technique converts a nominal input into an interval
input by using a function of the distribution of a target variable for each
level in the nominal input variable. SAS® Enterprise Miner™ has a facility to
create extension nodes that work in the same way as a usual node. This
paper explains creation of an extension node in SAS Enterprise Miner that
performs WOE recoding.
37
1:30 p.m.
4:30 p.m.
Take Home the ODS Crown Jewels: Master the New
Production Features of ODS LAYOUT and Report Writing
Interface Techniques
Extraction, Transformation, and Loading (ETL) for
Outcome Measures of Workers’ Compensation Benefits
Paper 015-2013
Paper 018-2013
The ODS “crown jewels” are the most sought-after features that allow the
customer to create sophisticated documents that can readily be published
via print, Web or mobile device. Journey with the SAS® 9.4 release as we
explore the new enhancements in this production release of ODS LAYOUT
and the Report Writing Interface. This example-driven content is intended
to empower and captivate the novice ODS customer while challenging
even the most advanced user.
Base SAS® was used to create a data sub-system for measuring outcomes,
added to a data system (coded in SAS) of benefit costs and employment.
One claim per injured worker per fiscal year is extracted as a study or
control record, using business-rule code. Disability benefits and
employment data are transformed to time-series records for claims, which
are transformed to time-series statistics by fiscal year. Programs are run
remotely on a UNIX data warehouse, and SAS data sets and metadata are
loaded to the warehouse and downloaded to a LAN. Quarterly generations
are kept for analysis of claim development.
Dan O'Connor, SAS
2:30 p.m.
Set Yourself Free: Use ODS Report Writing Technology in
SAS® Enterprise Guide® Instead of Dynamic Data
Exchange in PC SAS®, Part II SAS Code Revealed
Robert Springborn, Office of Statewide Health Planning &
Development
The ability to prepare custom designed reports and convey your message
in a clear and concise manner is very important in today’s sophisticated
business environment. Traditional use of Dynamic Data Exchange (DDE) in
PC SAS® to produce custom designed reports is the result of widespread
and popular use of Microsoft Excel. However with most business
organizations transitioning to SAS® Enterprise Business Intelligence (EBI),
where DDE is not compatible, ODS Report Writing technology is a powerful
alternative to create custom designed reports in SAS® Enterprise Guide®.
The driving force for this topic was the need to create hospital-level data
discrepancy reports which compare clinical data to administrative data to
verify risk factors used in a risk-adjusted operative mortality model
3:30 p.m.
Mike Maier, Oregon Department of Consumer and Business
Services
5:00 p.m.
Predictive Modeling in Sports Leagues: An Application in
the Indian Premier League
Pankush Kalgotra, Oklahoma State University
Ramesh Sharda, Oklahoma State University
Paper 019-2013
The purpose of this article is to develop models that can help team
selectors build talented teams with minimum possible spending. In this
study, we build several predictive models for predicting the selection of a
player in the Indian Premier League, a cricket league, based on each
player’s past performance. The models are developed using SAS® Enterprise
Miner™ 7.1. The best-performing model in the study is selected based on
the validation data misclassification rate. The selected model provides us
with the probability measure of the selection of each player, which can be
used as a valuation factor in the bidding equation. The models that are
developed can help decision-makers during auction set salaries for the
players.
The SAS® Output Delivery System: Boldly Take Your Web
Pages Where They Have Never Gone Before!
Paper 017-2013
8:00 a.m.
Chevell Parker, SAS
HTML output is one of the most effective and portable methods of
disseminating information within an organization. Using a variety of
techniques in the SAS® Output Delivery System (ODS), you can create HTML
output that increases the visibility and functionality of your web pages. This
paper discusses how to use those ODS techniques to deliver web content
specifically for mobile devices, web content for both mobile and desktop
devices, and web content specifically for desktop devices. In the first two
categories, the paper discusses the challenges and solutions for both types
of web content. For desktop devices, the paper discusses how to extract
data from web pages and place it into pivot tables for data-visualization
purposes in Microsoft Excel.
38
Be Prompt: Do it Now! Creating and Using Prompts in
SAS® Enterprise Guide®
Ben Cochran, The Bedford Group
Prompts are a quick and powerful way to give your programs, tasks, and
projects in SAS® Enterprise Guide® interactive capabilities. By putting
prompts in your code, you increase your ability to reuse code and also
enable the code to be customized using the value that is entered through
the prompt. Prompts are fairly easy to create, and this paper takes a stepby-step approach that explains how to create and use prompts.
9:00 a.m.
2:00 p.m.
Have it Your Way: Creating Reports with the DATA Step
Report Writing Interface
RUN_MACRO Run! With PROC FCMP and the
RUN_MACRO Function from SAS® 9.2, Your SAS®
Programs Are All Grown Up
Pete Lund, Looking Glass Analytics
SAS provides powerful, flexible reporting procedures. ODS provides
enormous control over the appearance of procedure output. Still, for times
where you need more, the Report Writing Interface can help. “Report
Writing Interface” simply refers to using the ODSOUT object in a DATA step.
This allows you to lay out the page, create tables, embed images, add titles,
and more using any desired DATA step logic. Most style capabilities of ODS
are available, so your output can have fonts, colors, backgrounds, and
borders to customize your report. This presentation will cover the basics of
the ODSOUT object and then walk through techniques to create four “real
world” examples. You might even go home and replace some PROC
REPORT code!
10:00 a.m.
Inventory Your Files Using SAS®
Brian Varney, Experis Business Analytics
Whether you are attempting to figure out what you have when preparing
for a migration or you just want to find out which files or directories are
taking up all of your space, SAS® is a great tool to inventory and report on
the files on your desktop or server. This paper presents SAS code to
inventory and report on the location you want to inventory.
11:00 a.m.
This Is the Modern World: Simple, Overlooked SAS®
Enhancements
Bruce Gilsen, Federal Reserve Board
Some smaller, less dramatic SAS® enhancements seem to fall through the
cracks. Users continue to employ older, more cumbersome methods when
simpler solutions are available. This includes enhancements introduced in
SAS 9.2, SAS 9, SAS 8, or even SAS 6! This paper reviews underutilized
enhancements that allow you to more easily 1. Write date values in the
form yyyymmdd. 2. Increment date values with the INTNX function. 3.
Create transport files: PROC CPORT/CIMPORT versus PROC COPY with the
XPORT engine. 4. Count the number of words or the number of occurrences
of a character or substring in a character string. 5. Concatenate character
strings. 6. Check if any of a list of variables contains a value. 7. Sort by the
numeric portion of character values. 8. Retrieve DB2 data on z/OS
mainframes.
1:30 p.m.
Submitting SAS® Code on the Side
Rick Langston, SAS
Paper 032-2013
Dylan Ellis, Mathematica Policy Research
Paper 033-2013
When SAS® first came into our life, it comprised but a DATA step and a few
procedures. Then we trained our fledgling programs using %MACRO and
%MEND statements, and they were able to follow scripted instructions. But
with SAS 9.2 and 9.3, your macros are now wearing the clothes of a PROC
FCMP function; you no longer need to feed every parameter with a spoon.
These functions are independent programming units, and this talk shows
how they can be put to use for handy calculations, standardizing and
simplifying code, and adding dynamic new capabilities that may change
the way you program.
2:30 p.m.
A Flock of C-Stats, or Efficiently Computing Multiple
Statistics for Hundreds of Variables
Steven Raimi, Magnify Analytic Solutions
Bruce Lund, Marketing Associates, LLC
Paper 034-2013
In other presentations, the authors have provided macros that efficiently
compute univariate statistics for hundreds of variables at a time. The classic
example is when a modeler must fit a binary model (two-valued target) and
has available hundreds of potential numeric predictors. Such situations may
occur when third-party data sets are added to in-house transactional data
for direct marketing or credit scoring applications. The paper describes the
SAS® code to compute these statistics, focusing on the techniques that
make these macros efficient. Topics include macro techniques for
identifying and managing the input variables, restructuring the incoming
data, and using hash objects to quickly count the number of distinct values
for each variable.
3:00 p.m.
A Better Way to Flip (Transpose) a SAS® Data Set
Arthur Tabachneck, myQNA, Inc.
Xia Keshan, Chinese Financial electrical company
Robert Virgile, Robert Virgile Associates, Inc.
Joe Whitehurst, High Impact Technologies
Paper 538-2013
Many SAS® programmers have flipped out when confronted with having to
flip (transpose) a SAS data set, especially if they had to transpose multiple
variables, needed transposed variables to be in a specific order, had a
mixture of character and numeric variables to transpose, or if they needed
to retain a number of non-transposed variables. Wouldn’t it be nice to have
a way to accomplish such tasks that was easier to understand and modify
than PROC TRANSPOSE, was less system resource-intensive, required fewer
steps, and could accomplish the task as much as fifty times or more faster?
This paper explains the new DOSUBL function and how it can submit SAS®
code to run “on the side” while your DATA step is still running. It also
explains how this function differs from invoking CALL EXECUTE or invoking
the RUN_COMPILE function of FCMP. Several examples are shown that
introduce new ways of writing SAS code.
39
3:30 p.m.
9:00 a.m.
Big Data, Fast Processing Speeds
Data Entry in SAS® Strategy Management: A New, Better
User (and Manager) Experience
Kevin McGowan, SAS
Paper 036-2013
David Shubert, SAS
As data sets continue to grow, it is important for programs to be written
very efficiently to make sure no time is wasted processing data. This paper
covers various techniques to speed up data processing time for very large
data sets or databases, including PROC SQL, data step, indexes and SAS®
macros. Some of these procedures may result in just a slight speed increase,
but when you process 500 million records per day, even a 10 percent
increase is very good. The paper includes actual time comparisons to
demonstrate the speed increases using the new techniques.
Paper 052-2013
9:30 a.m.
4:30 p.m.
Versatile Global Prompting for SAS® Web Report Studio
Hong Jiang, Deloitte
Maximizing the Power of Hash Tables
David Corliss, Magnify Analytic Solutions
Hash tables provide a powerful methodology for leveraging bid data by
formatting an n-dimensional array with a single, simple key. This
advancement has empowered SAS® programmers to compile exponentially
more missing data points than ever before, creating tables with hundreds
of fields of all types in which the majority of data in this vast array is empty.
However, the hash structure also supports analytics to calculate maximum
likelihood estimates for missing values, leveraging extensive data resources
available for each individual. An important application of this is in
sentiment analysis, where social media text expresses likes or dislikes for
particular products. Customer data, including sentiments for other
products, are used to model sentiment where an individual’s preference
has not been made known.
8:00 a.m.
SAS® BI Dashboard: Interactive, Data-Driven Dashboard
Applications Made Easy
Scott Sams, SAS
Paper 061-2013
The latest release of SAS® BI Dashboard gives you powerful new
functionality for designing dashboard applications, which are a set of
interactive dashboards that present a data-driven story to the user.
Previously, dashboards have been able to link to other applications and
pass parameters. Now with version 4.31 M2, dashboards can link to other
dashboards and pass the current click context as parameters, guiding your
users through their data with customized dashboard presentations. The
context parameters can even be passed to a stored process and have its
generated data presented by a custom dashboard. This paper shows you
how to build a simple dashboard application using key enhancements in
SAS BI Dashboard 4.31 M2.
40
Data entry in SAS® Strategy Management has never been an especially
pleasant task given the outdated user interface, lack of data validation and
limited workflow management. However, release 5.4 unveils a complete
overhaul of this system. The sleeker HTML5-based appearance provides a
more modern Web experience. You are now able to create custom data
validation rules and attach a rule to each data value. Discover how form
workflow is supported via SAS Workflow Studio and the SAS Workflow
Services.
Paper 054-2013
Prompts built into the information map are convenient tools for developers
using SAS® Web Report Studio. However, the parameter values set through
these prompts are not able to populate to other web report sections or
objects, limiting their usefulness. This paper describes a solution for
creating versatile global prompts that support one-time user response for
multiple objects or sections in a SAS web report. Both single-value and
multiple-value selection features can be implemented by following the
directions described in this paper.
10:00 a.m.
Big Data - Dream IT. Build IT. Realize IT.
Paul Kent, SAS
Andy Mendelsohn, Oracle Corporation
Maureen Chew, Oracle Corporation
Steven Holmes, Bureau of Labor Statistics
This session will present unique perspectives on building solutions for Big
Data architectures to enable turning the vision into reality. Guest speakers
Andrew Mendelsohn, Senior Vice President, Oracle Database Server
Technologies, and Paul Kent, SAS Vice President, Big Data, will discuss
collaborative efforts towards best-of-breed Big Data analytic solutions and
convergence of game-changing IT strategies. We'll also hear from the
Bureau of Labor & Statistics on how their SAS® usage produces one of the
mostly watched economic series (U.S. employment) each month.
11:00 a.m.
How to List All Users That Have Access to a SAS®
Information Delivery Portal 4.3 Page
Bernt Dingstad, If Insurance
Paper 055-2013
This paper describes how to access SAS® Metadata from a Base SAS® client
and make simple listings of often very urgent information and in the end
distribute this information utilizing the SAS Enterprise BI framework.
11:30 a.m.
Key Aspects to Implement A Perfect SAS® BI Platform
Interact between objects. - Drill and expand. - Collaborate. - Share results
on the Web and mobile devices. SAS Visual Analytics provides fast, effective
dashboards anywhere, regardless of how big your data may be.
Paper 489-2013
4:30 p.m.
A perfect SAS® architecture is not defined just by successful installation of a
SAS platform but also by ensuring good performance, easy maintenance,
compliance to all security, secured environment, scalability, good
administration practices, proper monitoring, seamless Integration with
Interfacing system, etc. SAS provides lot of flexibility in order to integrate
with other interfacing systems; however, a perfect SAS Enterprise
Implementation is not just driven with the maturity of SAS platform but
also requires matured implementation of other interfacing platforms. SAS
user experience starts from the first click on the SAS client and driven from
SAS environment capabilities and its integration to interfacing systems.
Hence interfacing systems also have a key role to play in order to get
perfect SAS architecture.
How to Automate Security Filters for SAS® OLAP Cubes
Using Users Groups Information Available in SAS®
Management Console
Gaurav Agrawal, Major Financial Company
1:30 p.m.
Whirlwind Tour Around SAS® Visual Analytics
Anand Chitale, SAS
Christopher Redpath, SAS
Paper 057-2013
SAS brings a revolutionary approach for analytical- based data visualization,
exploration and reporting. Come and join us for a whirlwind tour through
SAS® Visual Analytics 6.1, inspired by customer experiences and use cases.
We start with an overview of the technology concept behind SAS Visual
Analytics and enter into the Hub, leading to various destinations, including
the Data Builder, Explorer and Designer, ending the journey with Mobile
integration. We also offer a sneak peek into world of Microsoft Office. Don’t
worry; SAS coders are not left behind. The final stop is SAS® Enterprise
Guide®, where you learn how to leverage the technology behind SAS Visual
Analytics through SAS code.
2:30 p.m.
The Forest and the Trees: See It All with SAS® Visual
Analytics Explorer
Nascif Abousalh-Neto, SAS
Plinio Faria, Bradesco
Paper 048-2013
In order to limit the data that a user can access in an OLAP Cube, it is
required to use MDX conditions, and those expressions must be
customized for every OLAP Cube because each one has a different
structure. This paper will focus on showing how to automate the creation of
MDX conditions using the users groups information available in the
metadata server and in a way that is possible to change the data subset
that the users are allowed to see only changing the user group in SAS®
Management Console. It will be demonstrated also how to implement
OLAP Cube security adding directly the users login to a cube dimension.
5:00 p.m.
From Factbook to Dashboard in T Minus No Time and
Counting!
Alicia Betsinger, Office of Strategic Initiatives, UT System
Annette Royal, The University of Texas System
Paper 049-2013
The University of Texas System has been publishing detailed data on
institutional performance for years using static PDF files and Excel
documents. With requests for more data increasing, this approach was
unsustainable. The Office of Strategic Initiatives (OSI) was spending too
much time collecting and processing data for the Chancellor, Board of
Regents, and media. There was no time for in-depth research or analysis.
Instead of users using the data to help support better management, the
data was managing the users. What grew from an internal office’s need
morphed into a larger UT System need for a BI system that would support
the Chancellor’s framework by providing an accessible and customizable
tool for monitoring institutional performance and progress toward
transparency and accountability goals.
Paper 058-2013
It’s a jungle out there! A data jungle, that is. With so much data to process,
it’s too easy to get lost. What’s a data explorer to do? This paper explains
how data exploration journeys usually follow a generic workflow composed
of seven well-defined tasks that are easy to perform using SAS® Visual
Analytics Explorer. Armed with this knowledge, you will be able to see both
the forest and the trees, and never worry again about losing your way.
3:30 p.m.
Fast Dashboards Anywhere with SAS® Visual Analytics
Rick Styll, SAS
Paper 059-2013
Dashboards can be the best starting point to provide a high-level view of
the most relevant information to monitor, analyze and collaborate around
business performance. A growing number of organizations are creating
dashboards to improve fact-based decision making, but if the dashboards
fail to return results quickly or are difficult to understand, user adoption will
soon wane. SAS® Visual Analytics includes an authoring interface called
Designer that lets you build responsive, well-formatted and effective
dashboards that allow users to: - Place, size and style objects precisely. -
Data Management — Room 2001
8:00 a.m.
Data Fitness: A SAS® Macro-based Application for Data
Quality of Large Health Administrative Data
Mahmoud Azimaee, Institute for Clinical Evaluative Sciences
This paper introduces a SAS® macro-based application package as a
solution for creating automated data quality assurance reports for large
health administrative data. It includes methods and tools for developing
metadata for a SAS data holding, for measuring different data quality
indicators using a Data Quality Framework, and for generating automated
visual data quality reports. Because quality of data documentation should
be considered as a usability and interpretability factor for good quality data,
this application uses the same metadata developed for data quality
purposes to generate an automated web-based data dictionary as well.
41
9:00 a.m.
Adaptive In-Memory Analytics with Teradata: Extending
Your Existing Data Warehouse to Enable SAS® HighPerformance Analytics
Greg Otto, Teradata Corporation
Tom Weber, SAS
Paper 076-2013
SAS® High-Performance Analytics rapidly analyzes big data in-memory. The
Initial High-Performance Analytics SAS offering on Teradata co-locates SAS®
on the database nodes in a separate appliance. Data is replicated to the
appliance for use by the SAS analytics. SAS and Teradata have developed a
new in-memory analytics architecture that delivers the power of SAS HighPerformance Analytics to data in the Enterprise Data Warehouse, without
replication to an appliance. In this “Asymmetric” architecture, dedicated
SAS nodes access Teradata data on demand via high-speed networking.
Asymmetric in-memory processing extends a Teradata EDW to support SAS
High-Performance Analytics with minimal impact. This paper explains the
Asymmetric architecture and configuration options, and quantifies the
impact to Teradata systems that are extended to support SAS in-memory
analytics.
needed. This paper explains how SAS® Data Quality functions can be
invoked in database, eliminating the need to move data, thus delivering
data quality that meets the need for near-real-time performance for today’s
business. Graphic results comparing performance metrics of traditional
data quality operations against in-database data quality will be presented
along with details on how these results scale with database resources.
1:30 p.m.
The SQL Tuning Checklist: Making Slow Database
Queries a Thing of the Past
Tatyana Petrova, SAS
Jeff Bailey, SAS
Paper 080-2013
The DB2 SQL query had been running for 36 hours before it was killed.
Something had to be done, but what? Chances are, you’ve been there and
felt the pain and helplessness. Is there anything you can do to make it run
faster? Absolutely. This presentation teaches the steps and the mindset
required to solve this all-too-common problem. We cover specific
techniques that will enable you to solve the problem and discuss DB2,
Oracle, Greenplum and Teradata.
9:30 a.m.
2:30 p.m.
Best Practices in SAS® Data Management for Big Data
Need for Speed - Boost Performance in Data Processing
with SAS/Access® Interface to Oracle
Nancy Rausch, SAS
Paper 077-2013
This paper discusses capabilities and techniques for optimizing SAS® data
management products for big data. It demonstrates how SAS supports
emerging Apache Hadoop technologies, including details on the various
languages available in the platform and how to use them with SAS. An
overview of SAS Data Management integration with the SAS® LASR™
platform is also provided. Practical tips and techniques are presented,
including case studies that illustrate how to best use the SAS data
management feature set with big data.
10:30 a.m.
SAS® Data Integration Studio: The 30-Day Plan
Svein Erik Vralstad, Knowit Desicion Oslo AS
Paper 081-2013
Big Data is engulfing us. The expectations of users increase, and analytics is
getting more and more advanced. Timely data and fast results have never
had greater value. In data warehousing, analytics, data integration, and
reporting, there is an ever-growing need for speed. When operating in
environments where performance is of importance, it is of great value to
fully understand the interaction between the different components of the
environment. Hence, the importance of in-database execution is
accelerating. To know when to let SAS process data, and when to use
Oracle to perform the task is then of great value. This paper explores ways
to achieve substantial gains in performance when processing (read,
transform, calculate, and write) data in an effective manner.
John Heaton, Heritage Bank
3:00 p.m.
Paper 078-2013
Sharpening Your Skills in Reshaping Data: PROC
TRANSPOSE vs. Array Processing
When starting your journey with SAS® Data Integration Studio, it is
important to get the basics correct. This paper outlines the main framework
and activities needed within the first 30 days to set you up for success using
SAS Data Integration Studio.
11:00 a.m.
In-Database Data Quality: Performance for Big Data
Scott Gidley, SAS
Mike Frost, SAS
Charlotte Crain, SAS
Paper 079-2013
Data quality and high performance have joined forces. Today is an era of
big data, extremely large data warehouses and potential security issues for
moving data. Traditional data quality is performed with an ETL-like
operation of extracting, processing and publishing back to the source. To
guarantee high performance and assure data security, a new approach is
42
Arthur Li, City of Hope
A common data managing task for SAS® programmers is transposing data.
One of the reasons for performing data transformation is that different
statistical procedures require different data shapes. In SAS, two commonly
used methods for transposing data are using either the TRANPOSE
procedure or array processing in the DATA step. Using PROC TRANSPOSE
mainly requires grasping the syntax and recognizing how to apply different
statements and options in PROC TRANSPOSE to different types of data
transposition. On the other hand, utilizing array processing in the DATA
step requires programmers to understand how the DATA step processes
data during the DATA step execution. In this talk, these two methods will be
reviewed and compared through various examples.
4:00 p.m.
8:30 a.m.
Pulling Data from the Banner Operational Data Store
with SAS® Enterprise Guide: Not Only Fast but Fun!
Targeting Public Value in New Zealand
Michael O'Neil, Ministry of Social Development
Paper 083-2013
The New Zealand Ministry of Social Development is implementing what has
been called the “Investment-Based Approach,” which aims to improve
social sector performance through better targeting. Social and fiscal
outcomes can be better achieved through smart targeting using clientcentred evidence to inform strategic and case-level targeted decisions. This
paper describes progress to date.
Claudia McCann, East Carolina University College of Nursing
The assessment of learning and of services in higher education is crucial for
continued improvement. Administrators and faculty, therefore, require data
for their decision-making processes. There are many data input experts on
campus and, unfortunately, far fewer who can extract the data in the
aggregate form required by administrators, accreditors, and other
institutional stakeholders. The SAS® Enterprise Guide interface with the
Banner Operational Data Store is a very powerful combination of softwares
that enable the end user to quickly access the institution's data and
produce reports. More powerful still is the ability to bring other relational
databases, such as Excel spreadsheets, into the SAS Enterprise Guide
environment, thereby allowing variables not available in the Operational
Data Store to be used in comparative analyses.
4:30 p.m.
Best Practices in Enterprise Data Governance
Nancy Rausch, SAS
Scott Gidley, SAS
Paper 084-2013
Data governance combines the disciplines of data quality, data
management, data policy management, business process management
and risk management into a methodology that ensures important data
assets are formally managed throughout an enterprise. SAS has developed
a cohesive suite of technologies that can be used to implement efficient
and effective data governance initiatives, thereby improving an enterprise’s
overall data management efficiency. This paper discusses data governance
best practices. It explains where and how SAS® capabilities (such as the
business data network, reference data management, data federation, data
quality, data management and master data management) can be used to
ensure data governance initiatives remain successful, continue to deliver
overall return on investment, and gain buy-in across the enterprise.
8:00 a.m.
Information Retrieval in SAS®: The Power of Combining
Perl Regular Expressions and Hash Objects
Lingxiao Qi, Kaiser Permanente
Fagen Xie, Kaiser Permanente
Paper 091-2013
The volume of unstructured data is rapidly growing. Effectively extracting
information from huge amounts of unstructured data is a challenging task.
With the introduction of Perl regular expressions and hash objects in SAS®
9, the combination of these two features can be very powerful in
information retrieval. Perl regular expressions can be used to match and
manipulate various complex string patterns. The hash object provides an
efficient and convenient mechanism for quick data storage and retrieval. By
leveraging the best from both tools and applying it to the electronic
medical data, we show how pattern searching on free text is made easy
while reducing coding effort and increasing performance.
Paper 092-2013
9:00 a.m.
Using the Boosting Technique to Improve the Predictive
Power of a Credit Risk Model
Andres Gonzalez, Colpatria / Scotia Bank
Darwin Amezquita, Colpatria-Scotia Bank
Alejandro Correa Bahnsen, Luxembourg University
In developing a predictive model, the complexity of the population used to
build the model can lead to very weak scorecards when a traditional
technique such as logistic regression or an MLP neural network is used. For
these cases some nontraditional methodologies like boosting could help
improve the predictive power of any learning algorithm. The idea behind
this technique is to combine several weak classifiers to produce a much
more powerful model. In this paper, boosting methodology is used to
enhance the development of a credit risk scorecard in combination with
several different techniques, such as logistic regression, MLP neural
networks, and others, in order to compare the results of all methodologies
and determine in which cases the boosting algorithm increases model
performance.
9:30 a.m.
Using the Power of SAS® to Analyze and Solve Quality
Problems at Shanghai General Motors
Yu Zhang, Shanghai General Motors Co.,Ltd
Ying Wang, Shanghai General Motors Co.,Ltd
Shaozong Jiang, Information System Department,Shanghai
General Motors Co.,Ltd
Yahua Li, Information System Department,Shanghai General
Motors Co.,Ltd
Jiawen Zhang, Information System Department, Shanghai
General Motors Co.,Ltd
Nanxiang Gao, Quality Departmentï¼Shanghai General
Motors Co.,Ltd
Jian Li, Quality Departmentï¼Shanghai General Motors
Co.,Ltd
Minghua Pan, Quality Departmentï¼Shanghai General
Motors Co.,Ltd
Paper 090-2013
Data to assist in solving quality problems is of enormous value to quality
departments in the automotive industry, including that of Shanghai
General Motors (SGM). However, millions of claims records, tens of
thousands of solving reports, dozens of language descriptions, and
heterogeneous regional code present great difficulty for dataflow and
knowledge management. This paper explores SGM’s information system,
43
known as Problem Solving Analysis (PSA), which uses several foundation
tools of SAS®, such as Base SAS®, SAS/CONNECT®, and SAS/ACCESS®, to
solve business problems faster, and which has used advanced SAS®
Enterprise Content Categorization to establish 26,000 text rules for word
recognition and accurate classification. PSA incorporates effective data
infrastructure building, report linking, fast information searches, and
diagnosis and forecasting of enterprise problems.
10:00 a.m.
Creating Interval Target Scorecards with Credit Scoring
for SAS® Enterprise Miner™
Miguel Maldonado, SAS
Wendy Czika, SAS
Susan Haller, SAS
Naeem Siddiqi, SAS
Paper 094-2013
Credit Scoring for SAS® Enterprise Miner™ has been widely used to develop
binary target probability of default scorecards, which include scorecards for
application and behavior scoring. Extending the weight-of-evidence
binned scorecard methodology to interval-based targets such as loss given
default and exposure at default is a natural step in the evolution of
scorecard development methodologies. This paper discusses several new
methods available in Credit Scoring for SAS Enterprise Miner that help build
scorecards based on interval targets. These include cutoffs that convert the
interval target to a binary form and an algorithm that maintains the
continuous nature of the interval target. Each method is covered in detail,
and all methods are compared to determine how they perform against
known target values.
10:30 a.m.
Variable Reduction in SAS® by Using Weight of Evidence
and Information Value
Alec Lin, PayPal, a division of eBay
Paper 095-2013
Variable reduction is a necessary and crucial step in accelerating model
building without losing potential predictive power. This paper provides a
SAS® macro that computes weight of evidence and information value for all
potential predictors at the beginning stage of modeling. The SAS output
generated at the end of the program will rank continuous, ordinal, and
categorical variables by their predictive power, which can lend useful
insights to variable reduction. The reduced list of variables enables
statisticians to quickly identify the most informative variables for building
logistic regression models.
11:00 a.m.
Incremental Response Modeling Using SAS® Enterprise
Miner™
Taiyeong Lee, SAS
Ruiwen Zhang, SAS
Laura Ryan, SAS
Xiangxiang Meng, SAS
Paper 096-2013
Direct marketing campaigns that use conventional predictive models target
all customers who are likely to buy. This approach can lead to wasting
money on customers who will buy regardless of the marketing contact.
However, incremental response models that use a pair of training data sets
(treatment and control) measure the incremental effectiveness of direct
44
marketing. These models look for customers who are likely to buy or
respond positively to marketing campaigns when they are targeted but are
not likely to buy if they are not targeted. The revenue generated from those
customers is called incremental revenue. This paper shows how to find that
profitable customer group and how to maximize return on investment by
using SAS® Enterprise Miner™.
11:30 a.m.
Estimates of Personal Revenue from Credit and
Sociodemographic Information Combining Decision
Trees and Artificial Neural Networks (ANN)
Deybis Florez Hormiga, Colpatria Bank
Paper 097-2013
In different bank processes, a typical problem is determining customer
revenue. Customer revenue information is very important and highly
impacts these processes. As a result, finding a method to estimate the
revenue of customers for validation, segmentation, profiling, business
strategies, risk mitigation, regulatory compliance, or simply as information
is critical. Due to the amount of information and the high volatility of the
income reported by different clients, SEMMA methodology is used with
SAS® Enterprise Miner™. Starting from a fine segmentation using decision
trees and then Artificial Neural Networks (ANN) in each of the segments,
higher performance is achieved by including credit information and
customer sociodemographic variables.
1:30 p.m.
Where Should I Dig? What to Do before Mining Your
Data
Stephanie Thompson, Datamum
Paper 098-2013
Data mining involves large amounts of data from many sources. In order to
successfully extract knowledge from data, you need to do a bit of work
before running models. This paper covers selecting your target and data
preparation. You want to make sure you find gold nuggets and not pyrite.
The work done up front will make sure your panning yields results and is
not just a trip down an empty shaft.
2:00 p.m.
Relate, Retain, and Remodel: Creating and Using
Context-Sensitive Linguistic Features in Text Mining
Models
Russell Albright, SAS
Janardhana Punuru, SAS
Lane Surratt, SAS
Paper 100-2013
Text mining models routinely represent each document with a vector of
weighted term frequencies. This bag-of-words approach has many
strengths, one of which is representing the document in a compact form
that can be used by standard data mining tools. However, this approach
loses most of the contextual information that is conveyed in the
relationship of terms from the original document. This paper first teaches
you how to write pattern-matching rules in SAS® Enterprise Content
Categorization and then shows you how to apply these patterns as a
parsing step in SAS® Text Miner. It also provides examples that improve on
both supervised and unsupervised text mining models.
2:30 p.m.
Replacing Manual Coding of Customer Survey
Comments with Text Mining: A Story of Discovery with
Textual Data in the Public Sector
Jared Prins, Alberta Tourism, Parks and Recreation
A common approach to analyzing open-ended customer survey data is to
manually assign codes to text observations. Basic descriptive statistics of
the codes are then calculated. Subsequent reporting is an attempt to
explain customer opinions numerically. While this approach provides
numbers and percentages, it offers little insight. In fact, this method is
tedious and time-consuming and can even misinform decision makers. As
part of the Alberta government’s continual efforts to improve its
responsiveness to the public, the Alberta Parks division transitioned from
manual categorization of customer comments to a more automated
method that uses SAS® Text Miner. This switch allows for faster analysis of
unstructured data, and results become more reliable through the
consistent application of text mining.
3:30 p.m.
Data Mining of U.S. Patents: Research Trends of Major
Technology Companies
Ken Potter, SAIC
Robert Hatton, SAIC
Paper 101-2013
Research initiatives are normally closely held corporate secrets. Insights into
research trends are difficult to extract from public information, but data
mining of the U.S. Patent and Trademark Office (USPTO) patent grants
provides an opportunity to expose interesting trends and areas of interest
as indicated by activity in related patent areas. This paper covers assessing
the vast USPTO information repository and the analytical methodology that
extracts patent grant information from multiple formats and produces
interesting insights into research trends for several major technology
companies.
4:00 p.m.
A Tale of Two SAS® Technologies: Generating Maps of
Topical Coverage and Linkages in SAS User Conference
Papers
Denise Bedford, Kent State University
Richard La Valley, Strategic Technology Solutions
Barry deVille, SAS
Paper 102-2013
This paper discusses how SAS® technologies -- Text Analytics and Content
Categorization Suite -- were used to generate comprehensive and dynamic
summaries of the entire corpus of SAS user presentations from inception to
the present. The goal was to improve access to the conference proceedings
for SAS users and conference attendees in particular. The research
addresses two important access points to conference papers -- Industry
Solutions and Technology Solutions. The findings of this research suggest
that both suites are powerful tools that can be used in complementary or
independent approaches to generate similar results. The Industry Solution
perspective generated by both technologies surfaced common access
points. The Technology Solution perspectives also generated similar
perspectives when comparable rule sets were leveraged.
4:30 p.m.
Unleashing the Power of Unified Text Analytics to
Categorize Call Center Data
Saratendu Sethi, SAS
Jared Peterson, SAS
Arila Barnes, SAS
Paper 103-2013
Business analysts often want to take advantage of text analytics to analyze
unstructured data. With that in mind, SAS is delivering a new web-based
application that is designed to put the power of SAS® Text Analytics into
the hands of the analyst. This application brings together the power of SAS®
Text Miner and SAS® Content Categorization in a single user interface that
enables users to automatically create statistical and rule-based models
based on their domain knowledge. This paper demonstrates how a
business analyst in a call center environment can identify emerging topics,
generate automatic rules for those topics, edit and refine those rules to
improve results, derive insights through visualization, and deploy the
resulting model to score new data.
5:00 p.m.
Deciphering Emoticons for Text Analytics: A MacroBased Approach
Chad Atkinson, Sinclair Community College
Paper 104-2013
Emoticons, initialisms, and acronyms can evade routine processing, and the
difference between "this was a great class :)" and "ZOMG that was the best
class ever :-7" might be significant. This paper develops a macro that
converts select emoticons, initialisms, and acronyms to text that can be
parsed by SAS® Text Analytics or SAS® Sentiment Analysis Studio.
5:30 p.m.
Be Customer Wise or Otherwise: Combining Data Mining
and Interactive Visual Analytics to Analyze Large and
Complex Customer Resource Management (CRM) Data
Aditya Misra, Nanyang Technological University
Kam Tin Seong, Singapore Management University
Junyao Ji, SAS Institute
Paper 105-2013
In this competitive world, more and more companies, such as our project
sponsor, a global logistics company, are exploring the potential use of data
mining techniques to make informed and intelligent marketing strategies.
We conducted a market segmentation study using a comprehensive set of
customer transaction and profile data. This paper aims to report on our
experience gained in using the interactive visual analytics and data mining
techniques of JMP® to perform customer segmentation analysis. We share
our views on how interactive visual analytics and data mining techniques
can empower everyday data analysts to gain useful insights and formulate
informed decisions by demonstrating the interactive data visualization
techniques of JMP such as graph builder, parallel plots, and bubble plots.
45
Financial Services — Room 2010
1:30 p.m.
8:00 a.m.
How to Improve Your Risk-Adjusted Return on Capital:
Pricing Optimization in Lending
Macro Basics for New SAS® Users
Boaz Galinson, LEUMI
Lending is the core business of most banks. One may think that a bank
should opt for a lending pricing strategy of seeking the highest price that
the credit officer can obtain from his borrower. This paper claims that
following a "maximal price" strategy will eventually lead to an inferior credit
portfolio. I describe how to price a loan to meet at least the return required
by the stock holders and to improve RAROC. The strategy may be a
challenge in some assets classes. It can be difficult to agree on a price which
includes the minimal credit risk premium which compensates the risks. A
solution which accounts for all borrower activities with the bank is
presented.
2:30 p.m.
n Ounce of Prevention Is Worth a Pound of Cure: How
SAS® Helps Prevent Financial Crime with an Analytical
Approach to Customer Due Diligence
Scott Wilkins, SAS
Paper 107-2013
The increasing regulatory expectations on the risk rating of high-risk clients
and the emphasis on identification of foreign relationships with existing
customers has driven financial institutions to enhance their Customer Due
Diligence (CDD) processes. This paper outlines how organizations leverage
SAS® to deploy on-boarding and ongoing Customer Due Diligence
programs. It will explore analytical techniques for risk ranking customers,
best practices for deploying these programs, as well as how the SAS
approach incorporates a proactive, analytically driven triggering of new
investigations based on detected customer events or behavior. SAS can
provide organizations "The Power to Know" their customers and the risk
they may represent to their financial institutions.
3:30 p.m.
Next-Generation Detection Engine for Fraud and
Compliance
Ryan Schmiedl, SAS
Paper 108-2013
SAS’ next-generation approach provides a pivotal shift in how financial
institutions assess and govern customer risk. This paper discusses how
companies can aggregate, sum and understand patterns on huge volumes
of data; run more proactive what-if scenarios to identify and focus efforts
on the most critical investigations; and understand the impacts and
opportunity costs across scenarios.
46
Cynthia Zender, SAS
Paper 120-2013
Are you new to SAS®? Do you look at programs written by others and
wonder what those & and % signs mean? Are you reluctant to change code
that has macro variables in the program? Do you need to perform repetitive
programming tasks and don’t know when to use DO versus %DO? This
paper provides an overview of how the SAS macro facility works and how
you can make it work in your programs. Concrete examples answer these
and other questions: Where do the macro variables live? What does it mean
when I see multiple ampersands (&&)? What is a macro program, and how
does it differ from other SAS programs? What’s the big difference between
DO and IF and %DO and %IF?
9:00 a.m.
Reading Data from Microsoft Word Documents: It's
Easier Than You Might Think
John Bentley, Wells Fargo Bank
Paper 121-2013
SAS® provides the capability of reading data from a Microsoft Word
document, and it's easy once you know how to do it. Using the FILENAME
statement with the DDE engine makes it unnecessary to export to Excel or
work with an XML map. This paper goes through the steps of having SAS
read a Word document and shares a live example that demonstrates how
easy it is. All levels of SAS users may find this paper useful.
9:30 a.m.
Do Not Let a Bad Date Ruin Your Day
Lucheng Shao, University of California at Irvine
Paper 122-2013
Just as we have to step out of fairy-tale land and into reality when we grow
up, we can’t always expect the input dates to be good. This paper shows
you what SAS® does when it runs into input dates that are normally good
but have now gone bad, and how those problems can be addressed by
code. It is intended for readers who are familiar with Base SAS but not with
bad dates.
10:00 a.m.
PROC DATASETS: The Swiss Army Knife of SAS®
Procedures
This paper highlights many of the major capabilities of PROC DATASETS. It
discusses how it can be used as a tool to update variable information in a
SAS data set; provide information on data set and catalog contents; delete
data sets, catalogs, and indexes; repair damaged SAS data sets; rename
files; create and manage audit trails; add, delete, and modify passwords;
add and delete integrity constraints; and more. The paper contains
examples of the various uses of PROC DATASETS that programmers can cut
and paste into their own programs as a starting point. After reading this
paper, a SAS programmer will have practical knowledge of the many
different facets of this important SAS procedure.
11:00 a.m.
The SAS® Programmer's Guide to XML and Web Services
Paper 124-2013
Because of XML's growing role in data interchange, it is increasingly
important for SAS® programmers to become familiar with SAS technologies
and techniques for creating XML output, importing data from XML files, and
interacting with web services -- which commonly use XML file structures for
transmission of data requests and responses. The current work provides
detailed examples of techniques you can use to integrate these data into
your SAS solutions using SAS® XML Mapper, the XML LIBNAME engine, the
Output Delivery System, the FILENAME statement, and new SOAP functions
available beginning in SAS 9.3.
1:30 p.m.
Essentials of the Program Data Vector (PDV): Directing
the Aim to Understanding the DATA Step
Arthur Li, City of Hope
Beginning programmers often focus on learning syntax without
understanding how SAS® processes data during the compilation and
execution phases. SAS creates a new data set, one observation at a time,
from the program data vector (PDV). Understanding how and why each
automatic or user-defined variable is initialized and retained in the PDV is
essential for writing an accurate program. Among these variables, some
variables deserve special attention, including variables that are created in
the DATA step, by using the RETAIN or the SUM statement, and via BYgroup processing (FIRST.VARIABLE and LAST.VARIABLE). In this paper, you
are exposed to what happens in the PDV and how these variables are
retained from various applications.
2:30 p.m.
The Magnificent DO
Paul Dorfman, Dorfman Consulting
debuggers are good programmers. This paper covers common problems
including missing semicolons and character-to-numeric conversions, and
the tricky problem of a DATA step that runs without suspicious messages
but, nonetheless, produces the wrong results. For each problem, the
message is deciphered, possible causes are listed, and how to fix the
problem is explained.
4:30 p.m.
30 in 20 Things You May Not Know about SAS®
Tim Berryhill, Wells Fargo
Paper 128-2013
30 things you may not know SAS® can do. In 20 minutes, I hope to widen
your eyes and improve your programming. I have used SAS on many
platforms and operating systems, with many databases. Most of these ideas
will run anywhere SAS runs.
5:00 p.m.
Hashing in PROC FCMP to Enhance Your Productivity
Donald Erdman, SAS
Andrew Henrick, SAS
Stacey Christian, SAS
Paper 129-2013
Hashing has been around in the DATA step since 2002 starting with SAS® 9.
Hashing is used mainly to improve performance for activities like merging
and searching. Starting in SAS® 9.3, hashing functionality is now available in
user-defined subroutines through PROC FCMP. While subroutines already
encapsulate and modularize the code to make programs reusable, with the
addition of hashing, now users can extend the scope of their program and
tackle larger problems without sacrificing simplicity. The complete hashing
syntax supported in PROC FCMP will be outlined, as well as how it differs
from the DATA step. Examples will also be provided demonstrating just
how hashing in user-defined subroutines can be utilized to improve
performance and streamline an existing program.
Any high-level computer program can be written using just three
fundamental constructs: Sequence, Selection, and Repetition. The latter
forms the foundation of program automation, making it possible to execute
a group of instructions repeatedly, modifying them from iteration to
iteration. In SAS® language, explicit repetition is implemented as a standalone structural unit via the DO loop - a powerful construct laden with a
myriad of features. Many of them still remain overshadowed by the
tendency to structure code around the implied loop - even when it makes
the program more complex or error-prone. We will endeavor to both
straighten out some such incongruities and give the sense of depth and
breadth of the magnificent SAS construct known as the DO loop.
8:00 a.m.
3:30 p.m.
SAS® Workshop: SAS® Add-In for Microsoft Office 5.1
Eric Rossland, SAS
Paper 526-2013
This workshop provides hands-on experience using the SAS® Add-In for
Microsoft Office. Workshop participants will:
• access and analyze data
• create reports
• use the SAS add-in Quick Start Tools
Errors, Warnings, and Notes (Oh My): A Practical Guide
to Debugging SAS® Programs
Susan Slaughter, Avocet Solutions
Lora Delwiche, Univeristy of California, Davis
This paper is based on the belief that debugging your programs is not only
necessary, but also a good way to gain insight into how SAS® works. Once
you understand why you got an error, a warning, or a note, you'll be better
able to avoid problems in the future. In other words, people who are good
47
8:00 a.m.
• register a user in the metadata
• manage access to application features with roles
Introduction to ODS Graphics
10:00 a.m.
Chuck Kincaid, Experis Business Intelligence and Analytics
SAS® has a new set of graphics procedures called ODS Graphics. They are
built on the Graph Template Language (GTL) in order to make the powerful
GTL easily available to the user. PROC SGPLOT and PROC SGPANEL are two
of the procedures that can be used to produce powerful graphics that
previously required a lot of work. This upgrade is similar to the ease-of-use
upgrade in output manipulation when ODS was first published. This handson workshop teaches you how to use PROC SGPLOT and PROC SGPANEL
and the new capabilities they provide beyond the standard plot. By using
these new capabilities, anyone can tell the story better.
8:00 a.m.
How to Use ARRAYs and DO Loops: Do I DO OVER or Do I
DO i?
Jennifer Waller, Georgia Health Sciences University
Do you tend to copy DATA step code over and over and change the
variable name? Do you want to learn how to take those hundreds of lines of
code that do the same operation and reduce them to something more
efficient? Then come learn about ARRAY statements and DO loops,
powerful and efficient data manipulation tools. This workshop covers when
ARRAY statements and DO loops can and should be used, how to set up an
ARRAY statement with and without specifying the number of array
elements, and what type of DO loop is most appropriate to use within the
constraints of the task you want to perform. Additionally, you will learn how
to restructure your data set using ARRAY statements and DO loops.
9:00 a.m.
Some Techniques for Integrating SAS® Output with
Microsoft Excel Using Base SAS®
Vince DelGobbo, SAS
Paper 143-2013
This paper explains some techniques to integrate your SAS® output with
Microsoft Excel. The techniques that are presented in this paper require
only Base SAS® 9.1 or above software, and can be used regardless of the
platform on which SAS software is installed. You can even use them on a
mainframe! Creating and delivering your workbooks on demand and in real
time using SAS server technology is discussed. Although the title is similar
to previous papers by this author, this paper contains new and revised
material not previously presented.
10:00 a.m.
A Row Is a Row Is a Row, or Is It? A Hands-On Guide to
Transposing Data
Christianna Williams, Independent Consultant
Sometimes life would be easier for the busy SAS® programmer if
information stored across multiple rows were all accessible in one
observation, using additional columns to hold that data. Sometimes it
makes more sense to turn a short, wide data set into a long, skinny one—
convert columns into rows. Base SAS® provides two primary methods for
converting rows into columns or vice versa: PROC TRANSPOSE and the
DATA step. How do these methods work? Which is best suited to different
transposition problems? The purpose of this hands-on workshop is to
demonstrate various types of transpositions using the DATA step and to
unpack the TRANSPOSE procedure. Afterward, you should be the office goto gal/guy for reshaping data.
SAS® Workshop: SAS® Enterprise Guide® 5.1
Eric Rossland, SAS
Paper 527-2013
This workshop provides hands-on experience using SAS® Enterprise Guide®.
• access different types of data
• analyze data using the Data Explorer
• create reports and analyses
10:00 a.m.
SAS® Workshop: SAS® Platform Administration
Christine Vitron, SAS
Paper 528-2013
This workshop provides hands-on experience using SAS® Management
Console to administer the platform for SAS® Business Analytics. Workshop
participants will:
48
11:00 a.m.
Eric Rossland, SAS
Paper 529-2013
1:30 p.m.
• access different types of data
SAS® Workshop: Creating SAS® Stored Processes
• create reports and analyses
Eric Rossland, SAS
Paper 530-2013
This workshop provides hands-on experience creating SAS® Stored
Processes. Workshop participants will:
• use SAS® Enterprise Guide® to access and analyze data
• analyze data using the Data Explorer
3:30 p.m.
SAS® Visual Analytics 6.1
Eric Rossland, SAS
• create stored processes which can be shared across the organization
Paper 532-2013
• access the new stored process from the SAS® Add-In for Microsoft Office
1:30 p.m.
Know Thy Data: Techniques for Data Exploration
Andrew Kuligowski, HSN
Charu Shankar, SAS Institute Toronto
Get to know the #1 rule for data specialists: Know thy data. Is it clean? What
are the keys? Is it indexed? What about missing data, outliers, and so on?
Failure to understand these aspects of your data will result in a flawed
report, forecast, or model. In this hands-on workshop, you learn multiple
ways of looking at data and its characteristics. You learn to leverage PROC
MEANS and PROC FREQ to explore your data, and how to use PROC
CONTENTS and PROC DATASETS to explore attributes and determine
whether indexing is a good idea. And you learn to employ powerful PROC
SQL’s dictionary tables to explore and even change aspects of your data.
1:30 p.m.
3:30 p.m.
Hands-On SAS® Macro Programming Tips and
Techniques
The SAS® macro language is a powerful tool for extending the capabilities
of SAS. This hands-on workshop presents numerous tips and tricks related
to the construction of effective macros through the demonstration of a
collection of proven macro language coding techniques. Attendees learn
how to process statements containing macros; replace text strings with
macro variables; generate SAS code using macros; manipulate macro
variable values with macro functions; handle global and local variables;
construct arithmetic and logical expressions; interface the macro language
with the DATA step and SQL procedure; store and reuse macros;
troubleshoot and debug macros; and develop efficient and portable macro
language code.
Getting Started with the SAS/IML® Language
Rick Wicklin, SAS
Paper 144-2013
Do you need a statistic that is not computed by any SAS® procedure? Reach
for the SAS/IML® language! Many statistics are naturally expressed in terms
of matrices and vectors. For these, you need a matrix-vector language. This
hands-on workshop introduces the SAS/IML language to experienced SAS
programmers who are familiar with elementary linear algebra. The
workshop focuses on statements that create and manipulate matrices, read
and write data sets, and control the program flow. Learn how to write userdefined functions, interact with other SAS procedures and recognize
efficient programming techniques. Programs will be written using the SAS/
IML® Studio development environment. This course covers chapters 2-4 of
“Statistical Programming with SAS/IML Software” (Wicklin 2010).
2:30 p.m.
3:30 p.m.
Taking Full Advantage of sasCommunity.org: Your SAS®
Site
Don Henderson, Henderson Consulting Services
sasCommunity.org, a clearinghouse for technical information related to the
use of SAS® software, is managed and run by SAS users; free and open to all
SAS users to browse; contributed to by any SAS user once they create an ID;
and built on top of the same software as Wikipedia. So if you know how to
use Wikipedia, you have a head start on using sasCommunity.org. Learn
how to navigate the information contained on the site; discover the wealth
of its hidden treasures; and make even small contributions to enhance the
site for everyone. Find out the number of ways you can contribute, and
discover how you too can quickly make a difference in this worldwide
community of SAS users.
SAS® Workshop: SAS® Enterprise Guide® 5.1
Eric Rossland, SAS
Paper 531-2013
This workshop provides hands-on experience using SAS® Enterprise Guide®.
49
9:30 a.m.
4:30 p.m.
Estimating Patient Adherence to Medication with
Electronic Health Records Data and Pharmacy Claims
Combined
SAS® Workshop: SAS® Platform Administration
Christine Vitron, SAS
Paper 533-2013
This workshop provides hands-on experience using SAS® Management
Console to administer the platform for SAS® Business Analytics. Workshop
participants will:
• register a user in the metadata
• manage access to application features with roles
8:00 a.m.
Clinician Prescribing Feedback Site: Comparing Clinician
Prescribing Habits and Providing Actionable Patient
Lists
Michael Nash, Kaiser Permanente
Paper 165-2013
Which doctor is prescribing the most non-formulary medications? Which
patients are on a brand drug when an equivalent generic is available? These
questions and many more can be answered when using the Pharmacy
Feedback Site. This secure intranet site at Kaiser Northwest uses SAS/
GRAPH® HBAR and VBAR charts to compare clinician prescribing habits. Drill
down to compare all clinics, or all departments, or all doctors within a Clinic
or Specialty. Drill down even further to find patient lists so pharmacists or
clinician staff can perform outreach to members. The Pharmacy Feedback
Site also tracks costs and patients month to month. This paper shows you
how to create linked HTML files by using PROC GCHART and the HTML=
option.
8:30 a.m.
Identifying and Addressing Post-Marketing
Pharmaceutical Safety Surveillance and Spontaneous
Reported Events
Carrie Boorse, SAS
Kathy Schaan, SAS
Stuart Levine, SAS
Paper 166-2013
While pharmaceutical medications and medical devices must undergo
clinical trials to determine their safety, often their long-term side effects will
not be recognized until the medication or device has been approved and
consumed by a larger population of patients for a longer period of time.
Signals that an adverse event is brewing need to be identified by using
post-marketing data as expeditiously as possible. Using SAS® analytical
platform and foundation products, a solution was implemented that
incorporates recognized statistical procedures and enables users to
incorporate new and enhanced procedures for signal determination. The
solution also alerts users that a signal has been triggered, while also
managing the movements and results of the signal’s investigation.
50
Beinan Zhao, Palo Alto Medical Foundation Research
Insitute
Eric Wong, Palo Alto Medical Foundation Research Institute
Latha Palaniappan, Palo Alto Medical Foundation
Paper 167-2013
Estimating patient adherence to medication is critical for comparative
effectiveness, patient-centered outcomes research, and epidemiological
studies. Using a comprehensive electronic medical record system (EpicCare)
that has been in practice for 11 years with more than one million patients,
the prevailing adherence metrics (for example, medication possession ratio
and proportion days covered) were evaluated. However, these metrics
cannot be evaluated when a patient does not fill a medication order
(primary non-adherent) or fills it only once (early stop). With just a little
more effort, additional clinical information can be incorporated from
electronic health records to obtain refined estimates of adherence. This
paper proposes a few composite metrics that might be of specific interest
to researchers and clinicians.
10:00 a.m.
Measuring Medication Adherence with Simple Drug Use
and Medication Switching
Stacy Wang, Walgreens
Zhongwen Huang, Walgreens
Seth Traubenberg, Walgreen Co.
Paper 168-2013
In this paper, we demonstrate SAS®-based solutions that allow providers to
calculate adherence across a range of prescribing patterns. The code
provided allows PDC to be calculated at both the therapeutic class level
and the patient disease level. Refining existing methodologies has
increased the efficiency of the calculations.
10:30 a.m.
SAS® Tools for Transparent and Reproducible Research:
Medication History Estimator
Brian Sauer, SLC VA Medical Center
Tao He, University of Utah
Paper 169-2013
The Medication History Estimator (MHE) is designed to output data at the
course-level; i.e., one row per drug course. A course and period proportion
of days covered (PDC) is calculated for each medication. Reports that
describe the frequency and percent of users for each medication product,
average duration of medication courses, medication possession ratios and
Kaplan-Meier based persistency curves are automatically generated and
formatted for professional reports and journal publications.
11:00 a.m.
3:00 p.m.
What Do Your Consumer Habits Say About Your Health?
Using Third-Party Data to Predict Individual Health Risk
and Costs
Evaluating System-Wide Process Improvement in a
Health-Care System: Data Through Analysis
Albert Hopping, SAS
Satish Garla, SAS
Rick Monaco, SAS Institute
Sarah Rittman, SAS
Paper 170-2013
The Affordable Care Act is bringing dramatic changes to the health care
industry. Previously uninsured individuals are buying health insurance and
consuming health care services differently. These changes are forcing
insurers to reevaluate marketing, engagement and product design
strategies. The first step in addressing these challenges is understanding
the financial risk of new entrants into the marketplace. How do you predict
the risk of a person without any historical cost information? What if all you
know is the name and address? The finance industry has long been using
third-party consumer data to predict future finance habits and credit risk.
This paper takes a look at applying advanced analytics from SAS to thirdparty data for predicting health care utilization risk and costs.
1:30 p.m.
To Infinity and Beyond: Current and Future State of Big
Data and Analytics in Life Sciences
Matthew Becker, inVentiv Health Clinical
Our biggest asset is our data. We have all heard a semblance of these words
in the Life Sciences industry. The questions many of us ask are: Are we
tapping into the data as we should? Are we pulling the multiple avenues of
data together with all the parameters that could be analyzed? Are we
providing analytics in an educational way to our end-user(s)? In this
keynote, we will look at the current state of big data in the Life Sciences
industry and share a glimpse into the future of big data and analytics.
2:30 p.m.
Moving to SAS® Drug Development 4.2
Magnus Mengelbier, Limelogic
Eric Wong, Palo Alto Medical Foundation Research Institute
Lubna Qureshi, Palo Alto Medical Foundation Research
Institute
Pragati Kenkare, Palo Alto Medical Foundation Research
Institute
Dorothy Hung, Palo Alto Medical Foundation Research
Institute
Paper 172-2013
Disruptive system changes are required for sustaining high-quality and
affordable health-care delivery systems. Successful, transformative healthcare system changes are few and even fewer have been rigorously
evaluated. Electronic health records and changes in health IT provide an
opportunity to leverage an explosion of data in measuring the impact of
process improvement initiatives. This paper provides an example of
assessing the impact of a system-wide change in a large, multi-specialty
health-care system serving two million patients with a 13-year history of
using electronic health records. Lessons from ETL all the way to statistical
analysis are detailed including relevant SAS® procedures.
3:30 p.m.
Modern SAS® Programming: Using SAS® Grid Manager
and SAS® Enterprise Guide® in a Global Pharmaceutical
Environment
David Edwards, Amgen
Greg Nelson, ThotWave Technologies, LLC.
Amgen, like most large biotechnology companies, uses SAS® to support the
drug discovery process. Equipped with a vision to fully leverage its global
workforce and to maximize its IT investments, Amgen developed a research
informatics infrastructure based on SAS to deliver value around the globe.
This paper will highlight many aspects of this project including business
justification, requirements, design, verification and validation, and
production migration for over 1500 programmers and statisticians spread
across three continents. We will highlight some of the challenges we faced
and how these were overcome using improved processes, modern
technologies such as SAS® Grid Manager and SAS® Enterprise Guide® and
the combined efforts of a global project team.
Paper 171-2013
Life Science organizations have a long investment into business processes,
standards, and conventions that make it difficult to simply turn to a new
generation of analysis environments. SAS® Drug Development 4.1
integrates many key features found in current analysis environments that
are spread across several applications and systems, which need to be
monitored and managed accordingly. The paper considers a set of SAS®
programs and how the SAS Drug Development repository, workspace, and
workflow features support a common business process with all of the
associated tools and utilities. The result is a short list of points to consider
and some tricks for moving a business process from a PC SAS or SAS server
environment to the new release of SAS Drug Development.
4:30 p.m.
Medical Versus Pharmacy Insurance: Which Is More CostEffective for Providing the Prescription? Solving the
Problem Via SAS® Enterprise Guide®.
Amber Schmitz, Prime Therapeutics
Paper 174-2013
Data-driven decisions that provide strategic solutions. These are buzzwords
and phrases we have all heard before, but actually applying those words to
deliver actionable data is less commonplace than it should be. This paper
51
explores how to use both pharmacy and medical insurance claims data in
order to assess drug utilization behavior across the medical and pharmacy
insurance benefits. The final result provides actionable data to assess
moving drug fills to the more cost-effective insurance benefit. This paper
explores: 1) Writing programs for efficient data pulls, 2) Macrotizing
program code to allow for flexible analysis constraints, 3) Using SAS®
Enterprise Guide® Tasks for analysis, and 4) Demonstrating business
intelligence via built-in graph options.
5:00 p.m.
Employee Wellness Programming Using SAS® Enterprise
Guide®
Yehia Khalil, Norton Healthcare
Tina Hembree, Norton Healthcare
Sandra Brooks, Norton Healthcare
Paper 175-2013
More businesses are using employee wellness programs to improve the
health of their employees (improve productivity levels, reduce
absenteeism, and reduce disability claims) while at the same time reducing
health care costs. The success of any wellness program depends on two
main rudiments. One: identify factors that drive up health care costs in the
organization such as smoking, obesity, chronic conditions, and others. Two:
achieve adequate employee engagement level in wellness programs and
identify barriers to achieving this level. The real challenge for any wellness
program is to incorporate the different data sources such as health risk
assessments (HRAs), demographics, medical claims data, and focus group
reports to build a comprehensive wellness program that considers the
various needs of the business population.
9:00 a.m.
Managing and Monitoring Statistical Models
Nate Derby, Stakana Analytics
Managing and monitoring statistical models can present formidable
challenges when you have multiple models used by a team of analysts over
time. How can you efficiently ensure that you're always getting the best
results from your models? In this paper, we'll first examine these challenges
and how they can affect your results. We'll then look into solutions to those
challenges, including lifecycle management and performance monitoring.
Finally, we'll look into implementing these solutions both with an in-house
approach and with SAS® Model Manager.
10:00 a.m.
SAS® Certification: Understand the Benefits of SAS
Certification, Which SAS Certifications Are Available,
and What SAS Certification Can Do for You
Andrew Howell, ANJ Group Pty Ltd
Paper 191-2013
SAS® has long had certification available for its programming language and
for its flagship data mining product, SAS® Enterprise Miner™. More recently
with the release of the SAS®9 platform suite have come certifications in
SAS® Data Integration Studio, SAS® Business Intelligence, and SAS® Platform
Administration. But what are the benefits (and some of the misconceptions)
of SAS Certification? What is available, and what's in it for organizations,
their staff and for SAS consultants to become SAS Certified?
10:30 a.m.
8:00 a.m.
The Successful SAS® Shop: 10 Ideas, Suggestions, and
Radical Notions
Communicating Standards: A Code Review Experience
Paper 192-2013
David Scocca, Rho, Inc.
Paper 188-2013
We need ways to pass along good programming practices. All but the
smallest companies will have programmers with varying levels of tenure
and experience. Standards and best practices change, but in a deadlinedriven world, we re-use old programs with minimal revision. Programmers
develop habits and can be slow to incorporate new approaches that might
simplify code or improve performance. We developed and rolled out an inhouse code review process to address these issues. This paper reports our
strategy for promoting and performing the reviews and describes the
results.
8:30 a.m.
If You Have Programming Standards, Please Raise Your
Hand: An Everyman's Guide
Dianne Louise Rhodes, US Census Bureau
Paper 189-2013
This paper goes through a step-by-step process of developing
programming standards, classifying them, and entering them into a
database. This database can then be used to develop style sheets and check
lists for peer review and testing. Through peer reviews and in preparation
for them, programmers learn good programming practices. We describe in
detail the most common standards, and why and how they should be
applied.
52
Lisa Horwitz, SAS
A SAS® shop might have as few as one or two programmers or as many as
several thousand programmers, analysts, statisticians, data stewards,
administrators, and help desk monitors. These SAS shops might be very
new to their organizations, or they might have grown and evolved over a
period of many years. Regardless of their size, age, or overall mission, there
are some common factors that allow these groups of people to find
satisfaction and reward in what they do. This paper details 10 ideas,
suggestions, and radical notions to ensure happy and productive SAS
programmers.
11:30 a.m.
Creating an Interactive SAS® E-Textbook with iBooks
Author for the iPad
William Zupko, U. S. Census Bureau
Paper 193-2013
Mobile media is becoming more popular and prevalent in today's
workplace. Even though few apps on the iPad actively run SAS® programs,
the iPad can be utilized as a teaching tool and reference database. iBooks
Author allows for the creation of interactive textbooks from anybody that
allow users to learn SAS in a self-paced environment. Widgets allow
screenshots to show how programs are run and use reviews to check
comprehension. These widgets also allow the inclusion of Keynote slides.
These interactive textbooks are especially excellent for SAS conferences, as
the text can be applied directly and include PowerPoint presentations,
creating a mobile library that can be used in an easily accessible format,
perfect for reference or training.
8:00 a.m.
Automating the Flow of Presentations in Coder's Corner
or Quick Tips
Erik Tilanus, Synchrona
Paper 292-2013
The Quick Tips section (former Coder's Corner) is characterized by a rapid
flow of many short presentations. Reading bios and starting presentations
by hand is slowing down this flow. So we use SAS® to automate this flow.
8:15 a.m.
You’ve Got SASMAIL! A Simple SAS® Macro for Sending
e-Mails
Rajbir Chadha, Cognizant Technology Solutions
Paper 340-2013
This paper talks about a way to have SAS® send out automatic e-mails once
the process finishes and include the log output or reports as an attachment.
The SASMAIL custom macro function combines the functionality of
FILENAME EMAIL, SAS macros, and the SQL procedure to deliver the
intended results in a simplified way. The SASMAIL macro uses a lookup for
the user’s login and e-mail address. This allows the macro to work with
minimum or no arguments. The function allows users to customize what
they want to see in the e-mail, including the e-mail list, the attachment, and
even the e-mail content. Users can even include a Microsoft Excel file or a
summary report as the attachment.
8:45 a.m.
Adding Graph Visualization on SAS® ODS Output
Miriam McGaugh, Oklahoma State University
Paper 310-2013
SAS® tools are normally used to produce statistical graphs such as pie
charts, bar charts, various plots, dashboards, and even geographical maps.
However, many SAS users may want to enhance their output by
incorporating various diagrams such as networking, cluster, and process
flows. In this paper, we will introduce a method to add specific graphs onto
SAS ODS output by interacting with Graphviz (an open source graph
visualization software) in Base SAS®.
9:00 a.m.
"How May I Help?" The SAS® Enterprise Guide® Analyze
Program Feature
Paper 311-2013
Have you ever been looking into a lengthy and complex SAS® code, maybe
inherited it or your own old program, and wished you could understand
what is happening inside without having to go through every line? Left
without any associated documentation and wondered where to start from?
Felt that it would be better to have a process flow representation of what
the code does, quickly? Then, the Analyze Program feature that SAS®
Enterprise Guide® offers might work for you! This paper discusses what to
expect of this feature and what not to with example analyses from the Life
Sciences industry.
9:15 a.m.
8:30 a.m.
Some Useful Utilities on UNIX Platform
Reporting Tips for No Observations
Paper 312-2013
Wuong Jodi Auyuen, Blue Cross Blue Shield Minnesota
Paper 320-2013
The goals for SAS developers to design applications include reporting
accurate information, delivering in a timely manner, meeting business
needs, and the presentation is easy to grasp. We design the report to meet
those goals and hopefully to cover potential questions. One of the
frequently asked questions is: I used to receive a session, say visitors from
Japan, why I don't see that session for the week of March 14, 2011? Even
though we don't need to code "No visitors from Japan due to Tsunami on
March 11, 2011", we could at least provide a generic message like "No data
returned for this session." so users won't be wondering whether they miss
the page or question the accuracy of the development work.
Kevin Chung, Fannie Mae
While using SAS® in UNIX platform, you might want to quickly browse data,
contents or the frequency of characteristic data fields in a SAS data set. You
can always write a SAS program and submit the program to get the results
you need. However, we are able to obtain this information in more efficient
and effective manner by using the UNIX shell scripts along with SAS codes.
This paper demonstrates some useful utilities in UNIX. This approach not
only saves your time but it also increases the productivity.
9:30 a.m.
PC and UNIX SAS® Reunited
Shiva Kalidindi, Amgen
Sarwanjeet Singh, Gerard Groups Inc.
Paper 313-2013
Have you ever wondered how you can use the best of PC and UNIX SAS
together and make a perfect world (well, almost perfect)? SAS/CONNECT®
allows you to use the Enhanced Editor and submit the code on UNIX. You
can submit one DATA step or PROC at a time, view the log in the Log
window as well as create data sets in the Work directory. It is a one-time
setup, and you do not have to compromise the PC Enhanced Editor ever.
This paper provides step-by-step instructions on how you can connect and
automate the PC-to-UNIX connection by using SAS. You will never have to
leave the Enhanced Editor again.
53
9:45 a.m.
10:45 a.m.
SAS® Code to Make Excel Files Section 508 Compliant
Reordering Columns after PROC TRANSPOSE (or
Anytime You Want, Really)
Christopher Boniface, U.S. Census Bureau
Hung Pham, U.S. Census Bureau
Nora Szeto, U.S. Census Bureau
Paper 314-2013
Can you create hundreds of great looking Excel tables all within SAS® and
make them all Section 508 compliant at the same time? This paper will
examine how to use ODS tagsets, EXCELXP, and other Base SAS® features to
create fantastic-looking Excel worksheet tables that are all Section 508
compliant. This paper will demonstrate that there is no need for any
outside intervention or pre- or post-meddling with the Excel files to make
them Section 508 compliant. We do it all with simple Base SAS code.
10:00 a.m.
Reading an Excel Spreadsheet with Cells Containing Line
Endings
Larry Hoyle, IPSR, University of Kansas
Paper 315-2013
The creative ways people enter data into Excel spreadsheets can cause
problems when trying to import data into SAS® data sets. This paper
addresses the problem encountered when spreadsheet cells contain
multiple lines (that is, the cells have embedded line endings). Several
approaches to reading such data are described and compared.
10:15 a.m.
Maintaining Formats When Exporting Data from SAS®
into Microsoft Excel
Nate Derby, Stakana Analytics
Colleen McGahan, BC Cancer Agency
Paper 316-2013
Data formats often get lost when exporting from SAS® into Microsoft Excel
using common techniques such as PROC EXPORT or the ExcelXP tag set. In
this paper, we describe some tricks to retain those formats.
10:30 a.m.
Don’t Let the Number of Columns Hold You Back!
Douglas Liming, SAS
Paper 318-2013
Many databases have a column limit of approximately 2,000 columns.
Several SAS® PROCs, such as PROC NEURAL and PROC TRANSPOSE, produce
output that easily exceeds 2,000 columns. Here is a technique to code
around this business problem when using databases on the backend.
Maximize your columns using SAS to talk to the databases via multiple
tables, pull them together and split them back out.
Sau Yiu, Kaiser Permanente
Paper 319-2013
There are times when we want to rearrange the order of the columns in a
SAS® data set. This occurs most often after a PROC TRANSPOSE, when the
newly transposed columns do not appear in the order that we want. This
paper shows several methods which allow users to either sort the columns
by their names, or order the columns in any particular way.
11:00 a.m.
Wide-to-Tall: A Macro to Automatically Transpose Wide
SAS® Data into Tall SAS Data
James R Brown, Havi Global Solutions
Paper 321-2013
If your SAS® world involves forecasting or other date-specific data, you have
probably seen column names such as forecast_19224, sales_19230, or
inventory_19250. If several of these prefixes exist in a single file, the
underlying SAS data file could have thousands of columns. Analyzing this
data is an exercise in scrolling, note-taking, copying, and pasting. PROC
TRANSPOSE is not sophisticated enough to take on this challenge. This
paper presents a macro which will transform your data by automatically
creating a CSV file with distinct columns for the date, each prefix variable,
and any non-date-suffixed columns in your input. The non-wizardry behind
this makes use of the dictionary tables, SAS name lists (forecast_18950forecast_19049), and colon notation (forecast_:) to eliminate the task of
enumerating long lists of variable names.
11:15 a.m.
Nifty Tips For Data Change Tracking
Julie Kilburn, City of Hope
Rebecca Ottesen, City of Hope and Cal Poly State University,
San Luis Obispo
Paper 333-2013
Best practices for databases include keeping detailed audit trail information
about the data. These audit trail tables vary in complexity as well as size.
Generally speaking, the larger the database in tables (as well as in
observations), the larger the audit trail. We have discovered that leveraging
audit trail information in our automated reporting has been a huge
resource saver in terms of which observations need to be reprocessed for a
report. Even with minimal audit information (such as created by and
modified by dates at the data table level), automation processing time can
be greatly reduced by taking advantage of a new way of thinking and a few
handy SAS® functions.
11:30 a.m.
Programs? How to Process Your Inputs Faster
Jason Wachsmuth, Pearson
Paper 334-2013
This paper demonstrates how to pass multiple input files as macro variables
and run multiple SAS® programs in one batch. This technique uses
%INCLUDE statements, CALL SYMPUT, and SCAN functions in a control
program to avoid physically opening, running, and closing each program.
Implementing this style of programming replaces the bottleneck of
54
defining %LET statements and enables you to process input files and
sequence dependent programs in proper order. Anyone who processes
data for a variety of routine operations will appreciate this solution.
efficiency improvement. This paper shows some sample code that divides
by zero and some benchmark results from changing the code to test for
zero denominators to avoid dividing by zero.
1:30 p.m.
2:30 p.m.
What's in a SAS® Variable? Get Answers with a V!
Running SAS® Programs Using Skype
Paper 322-2013
Paper 304-2013
If you need information on variable attributes, PROC CONTENTS will
provide all of the specifics. You could also access the SQL DICTIONARY
which contains tables filled with details on the variables in the active data
sets. If you need just one piece of information on a single variable, both of
these methods could prove to be cumbersome. However, SAS® has a whole
series of functions that can produce the information on one attribute of one
variable at one time. These functions are named with a V prefix followed by
descriptive term for the attribute. They include VNAME, VLABEL, VTYPE,
VFORMAT, and several others. We will demonstrate how these V functions
are useful not only in reporting but in standardizing the structure of a
database.
Skype is a well-known method used to talk to friends, family members, and
coworkers. It is one of the best applications available to make voice calls
over the Internet. In this paper we present a new, innovative way to use
SAS® with Skype. Here, we have developed a solution that allows users to
run SAS remotely through Skype. After installing the DLL from the API on
the application website, programmers can create scripts to control Skype.
By sending a specific message to a predefined user, programmers can
execute SAS on demand. This paper explains how to use Skype to run SAS
programs. It provides the Visual Basic script needed to communicate with
Skype and illustrates a real case scenario in which this technique is used.
William Murphy, Howard M Proskin & Assoc, Inc
1:45 p.m.
Quick and Easy Techniques for Fast Data Extraction
Mythili Rajamani, Kaiser Permanente
Deepa Sarkar, Kaiser Permanente
Jason Yang, Kaiser Permanente
Chris Greni, Kaiser Permanente
Paper 323-2013
Working with large data sets is a challenging and time-consuming job. This
paper tells some of the easy, useful data extraction tips and techniques to
reduce CPU usage to retrieve the specific data. The following subjects are
discussed in this paper: (1) creating a temporary table with the Key column
in the database (both DB2 and Teradata) and extracting the data from the
database (2) extracting data only for specific days, specific weeks (3)
automating the date parameter for repetitive and scheduled tasks.
Romain Miralles, Genomic Health
2:45 p.m.
An Overview of Syntax Check Mode and Why It Is
Important
Thomas Billings, Union Bank
Paper 327-2013
The syntax check options direct the SAS® system, when a syntax error
occurs while compiling source code, to enter a special mode to scan the
remainder of the job after the point where the error occurred, for syntax
errors. In this mode, only the header portion of some data sets are created,
permanent data sets are not replaced, but global commands are executed
(also a very few PROCs). The options controlling the mode are explained
and illustrated using simple test jobs. The effects of setting and resetting
the option within a job are explored, and there are some surprises along
the way. The risks of running with the options enabled vs. disabled are
discussed.
2:00 p.m.
3:00 p.m.
Dealing with Duplicates
Converting Thousands of Variables from Character to
Numeric: The One-Hour Fix
Christopher Bost, MDRC
Paper 324-2013
Variable values might be repeated across observations. If a variable is an
identifier, it is important to determine whether values are duplicated. This
paper reviews techniques for detecting duplicates with PROC SQL,
summarizing duplicates with PROC FREQ, and outputting duplicates with
PROC SORT.
2:15 p.m.
Not Dividing by Zero: The Last of the Low-Hanging
Efficiency Fruit
Bruce Gilsen, Federal Reserve Board
Wen Song, ICF International
Kamya Khanna, ICF International
Paper 328-2013
At the conclusion of many survey-based data collecting projects, recoding
the hundreds and thousands of character variables to “reserved scale”
specified numeric variables is a uncomplicated but cumbersome task for
SAS® programmers. If you are a person who likes to avoid a large amount of
typing as much as I do, this paper will give you an idea of how to maintain
high quality for this recoding task with minimal typing. This paper also
answers the following questions: How can you create a powerful SAS macro
that will write the IF-ELSE-THEN Statement for you? How can you avoid any
human errors such as typos? And how do you use Microsoft Excel to speed
up your work?
Paper 325-2013
As SAS® Institute has improved the efficiency of its code, some of the old
ways for users to improve efficiency, such as using WHERE or WHERE=
instead of IF in the DATA step, no longer make much difference. Current
user efforts to improve efficiency tend to focus on more sophisticated
techniques such as indexes or hashing. However, one of the classic
methods, not dividing by zero in the DATA step, can still offer a large
55
3:15 p.m.
4:15 p.m.
Resources for Getting the 2010 US Census Summary
Files into SAS®
Checking Out Your Dates with SAS®
San Luis Obispo
Paper 329-2013
At first glance, accessing the 2010 US Census data with SAS® seems like a
daunting task. The main limitation is that for the 2010 summary files it
seems that the Census has gravitated toward supporting data access via
Microsoft Access rather than SAS as they did in the past. However, there are
several tactics that can be deployed to make accessing this data with SAS
much easier. A thorough understanding of the Census Summary File data
structure and documentation can be used to leverage both SAS code from
programs that the Census previously supported and Census 2010 versioned
SAS programming available through other public sources. Knowledge of
the available resources can assist SAS analysts in taking advantage of this
rich data set.
3:30 p.m.
Array, Hurray, Array: Consolidate or Expand Your Input
Data Stream Using Arrays
William Benjamin, Owl Computer Consultancy LLC
Paper 330-2013
You have an input file with one record per month, but need a file with one
record per year. But you cannot use PROC TRANSPOSE because other fields
need to be retained or the input file is sparsely populated. The techniques
shown here enable you to either consolidate or expand your output data
using arrays. Sorted files of data records can be processed as a unit using
"BY Variable" groups and building an array of records to process. This
technique allows access to all of the data records for a "BY Variable" group
and gives the programmer access to the first, last, and all records in
between at the same time. This will allow the selection of any data value for
the final output record.
Christopher Bost, MDRC
Paper 335-2013
Checking the quality of date variables can be a challenge. PROC FREQ is
impractical with a large number of dates. PROC MEANS calculates summary
statistics but displays results as SAS® date values. PROC TABULATE,
however, can calculate summary statistics and format the results as dates.
This paper reviews these approaches plus the STACKODS option in SAS® 9.3
that might make PROC MEANS the preferred method for checking out your
dates.
4:30 p.m.
We All Have Bad Dates Once in a While...
Randall Deaton, BlueCross BlueShield of Tennessee
Patrick Kelly, BlueCross BlueShield of Tennnessee
Paper 336-2013
Dates in a corporate data arena can be a dangerous liaison. The strain of
translating corporate date types to SAS® date types can be tricky to
navigate, let alone bringing a third-party date type into the mix. Adding
third wheel to your SAS dates can create a comedy of errors. An
experienced SAS programmer with knowledge of the SAS macros and a few
clever programming tricks can more easily resolve your SAS Dates from a
sticky situation to an orderly affair.
4:45 p.m.
Increase Your Productivity by Doing Less
Robert Virgile, Robert Virgile Associates, Inc.
Arthur Tabachneck, myQNA, Inc.
Xia Keshan, Chinese Financial electrical company
Joe Whitehurst, High Impact Technologies
Paper 517-2013
3:45 p.m.
10-Minute JMP®
George Hurley, The Hershey Company
Paper 331-2013
Heard of JMP®, but haven't had time to try it? Don't want to devote 50
minutes to a talk about software that you might not want to use? This is the
talk to you. In 10 minutes, you will learn some of the amazing visualization
and modeling features in JMP and how to use them. This talk will JMP-start
your JMP usage. When it's complete, we suspect you will want to attend
some of the longer talks, too.
4:00 p.m.
Cool Views
Elizabeth Axelrod, Abt Associates Inc.
Paper 332-2013
Looking for a handy technique to have in your toolkit? Consider SAS® Views,
especially if you work with large data sets. After a brief introduction to
Views, I will show you several cool ways to use them that will streamline
your code and save workspace.
56
Using a keep dataset’ option when declaring a data option has mixed
results with various SAS procedures. It might have no observable effect
when running PROC MEANS or PROC FREQ but, if your datasets have many
variables, it could drastically reduce the time required to run some procs
like PROC SORT and PROC TRANSPOSE. This paper describes a fairly simple
macro that could easily be modified to use with any proc that defines which
variables should be kept and, as a result, make your programs run 12 to 15
times faster.
2002
8:00 a.m.
Make a Good Graph
Sanjay Matange, SAS
Paper 361-2013
A graph is considered effective if the information contained in it can be
decoded quickly, accurately and without distractions. Rules for effective
graphics – developed by industry thought leaders such as Tufte, Cleveland
and Robbins – include maximizing data ink, removing chart junk, reducing
noise and clutter, and simplifying the graph. This presentation covers these
principles and goes beyond the basics, discussing other features that make
a good graph: the use of proximity for magnitude comparisons, nonlinear
9:00 a.m.
the names of the style attributes that you want to change. This presentation
provides concrete examples to illustrate how to use STYLE= overrides with
PRINT, REPORT, and TABULATE. As the examples move from simple to
complex, you learn how to change fonts, add text decoration, alter the
interior table lines, perform traffic-lighting, and insert images into your ODS
output files using some ODS magic to improve your reports.
A Beginner's Introduction to an Idiot's Guide to PROC
TEMPLATE and GTL for Dummies
1:30 p.m.
or broken axes, small multiples, and reduction of eye movement for easier
decoding of the data. We also examine ways in which information can be
obscured or misrepresented in a graph.
Paper 363-2013
The aim of this paper will be an extremely gentle introduction to the very
exciting and somewhat intimidating world of PROC TEMPLATE and Graph
Template Language (GTL). As these two SAS® features are still relatively
new, not many people have had time to learn to learn them and see what
they are capable of accomplishing with minimal effort.
9:30 a.m.
Creating a Useful Operational Report
Andrew Hummel, Delta Air Lines
Robert Goldman, Delta Air Lines
Paper 364-2013
Metrics and reports are highly valued by operational decision makers in
order to make informed and data driven conclusions. However, there is a
balance between presenting useful organized knowledge and displaying
page upon page of raw useless data. SAS® has the power to produce a wide
range of sophisticated graphs that are meaningful. The challenge is to
produce a graph that quickly and accurately measures and displays the
real-world operation while allowing decision makers to make operationally
beneficial determinations. There are numerous SAS papers that give stepby-step instructions on how to build a graph; this is not such a paper. The
goal of this paper is to show how we approached the challenge of building
an operational report and the techniques used.
10:00 a.m.
Cascading Style Sheets: Breaking Out of the Box of ODS
Styles
Kevin Smith, SAS
Paper 365-2013
While CSS has been available in various forms in ODS since SAS® 9.2, SAS 9.4
is the first version that fully utilizes the new style engine’s architecture.
Using the new CSS engine, it is possible to apply custom styles based on
column names, BY group names, BY group variables and anchor names. It is
also possible to specify dynamic attribute values using symget, resolve,
dynamic and expression just like in PROC TEMPLATE styles. If you want to
break out of the box of ODS styles and do some truly original styling, the
CSS techniques in this paper will take you there.
Go Mobile with the ODS EPUB Destination
David Kelley, SAS
Julianna Langston, SAS
Paper 368-2013
The Base SAS® Output Delivery System (ODS) makes it easy to generate
reports for viewing on desktops. What about mobile devices? If you need
on-the-go reports, then the new SAS 9.4 ODS EPUB destination is the ticket.
With ODS EPUB, you can generate your reports as e-books that you can
read with iBooks® on the iPad®, or you can write an e-book from scratch.
This paper provides an introduction to writing e-books with ODS EPUB.
Please bring your iPad, iPhone® or iPod® so that you can download and read
the examples.
2:30 p.m.
Using Design Principles to Make ODS Template
Decisions
Helen Smith, RTI International
Susan Myers, RTI International
Paper 369-2013
With the Output Delivery System (ODS), SAS® continues to provide
programmers with many style templates for developing reports. These
default templates and style definitions present the data in a clear and
attractive manner often with no further thought needed. However, when
producing complicated reports with multiple requirements, using basic
design principles to determine which template or which custom style
definition to use can make for a more readable and comprehensive final
report. This paper presents the code and the design considerations for two
ODS reports; one, a redesign of a 10-plus-year-old SAS program originally
designed with PUT statements, and two, a highly customized SAS program
for delivering output in Microsoft Excel.
3:00 p.m.
Extended SAS® GIFANIM Device Usage on Table
Reporting and Template-Based Graphics
Xin Zhang, Emory University
Neeta Shenvi, Emory University
Azhar Nizam, Emory University
Paper 375-2013
11:00 a.m.
Turn Your Plain Report into a Painted Report Using ODS
Styles
Cynthia Zender, SAS
Allison Booth, SAS
Paper 366-2013
To use STYLE= statement level overrides, you have to understand what
pieces or areas of PRINT, REPORT, and TABULATE output you can change.
And then you have to understand how and where in your procedure syntax
you use the STYLE= override syntax. Last, but not least, you have to know
Dynamic, rather than static, graphs and tables often are more effective,
interactive, and audience-engaging presentations. The SAS® GIFANIM
device enables analysts to create GIF file-based slide shows for web and
PowerPoint presentations, but it only supports device-based graphics and
does not support SG procedure graphics. The GIFANIM device does not
provide an animation of summary tables. In this paper, several ways of
animating stand-alone summary tables, SG procedure graphics, and
graphics with embedded tables, using combinations of the SAS DATA Step
Graphics Interface (DSGI), printer-based methods, and Annotate data sets
are explored. The advantages and disadvantages of each method are
evaluated.
57
3:30 p.m.
Retail — Room 3014
Free Expressions and Other GTL Tips
8:00 a.m.
Prashant Hebbar, SAS
Sanjay Matange, SAS
Paper 371-2013
The Graph Template Language (GTL) provides many powerful features for
creating versatile graphs. The Statistical Graphics Engine in GTL provides
novel ways of using expressions to simplify the task of data preparation.
This presentation covers some new ways of using DATA step functions to
create grouped effect plots based on conditions and to select a subset of
the observations. It also illustrates using PROC FCMP functions in GTL
expressions. Novel uses of non-breaking space for creating small-multiples
graphs and graphs with indented text are discussed. Learn how to express
yourself with ease, graphically!
4:30 p.m.
Analysis and Visualization of E-mail Communication
Using Graph Template Language
Atul Kachare, SAS
Paper 372-2013
Email plays an important role in the corporate world as a means for
collaboration and sharing knowledge. Analyzing email information reveals
useful behavioral patterns that usually carry implicit information regarding
the senders common activities and interests. This paper demonstrates how
we can use SAS® data processing capabilities to extract vital information
based on corporate email communication data by reading email data and
graphically visualizing different communication patterns using Graph
Template Language. This analysis reveals different usage patterns: timebased email volumes with milestones; email exchanges within and across
groups; and communication preferences such as small emails, image-heavy
emails and the words most commonly used in communication.
5:00 p.m.
Visual Techniques for Problem Solving and Debugging
Who Said Change Was Easy
Scott Sanders, Sears
Allan Beaver, Soebeys
Margaret Pelan, Hudson Bay Company
Marty Anderson, Belk
In the modern business environment, organizations face rapid change like
never before. Due to the growth of technology, modern organizational
change is largely motivated by exterior innovations rather than internal
moves. When these developments occur, the organizations that adapt
quickest create a competitive advantage for themselves, while the
companies that refuse to change get left behind. Hear how 3 companies
have dealt with Change management and the lessons learned!
9:30 a.m.
SAS® Visual Analytics and Mobile Reporting
Frank Nauta, SAS
Paper 389-2013
Retailers are striving for an omnichannel, customer-centric experience with
brand consistency across all available touch points. It is vital to understand
customers well enough to anticipate their behaviors, know their
preferences and predict when those behaviors and preferences will change.
Data is key to gaining this level of insight. The sheer volume and complexity
of retail data can be a challenge, as can the inability to determine which
variables are relevant to your business. SAS Visual Analytics changes this
with literally a touch of the mouse! This session demonstrates analyzing
hundreds of millions of retail transactions so you can derive insights on ALL
your data in matter of a few seconds and distribute that info to users in an
easy-to-consume manner.
10:30 a.m.
Andrew Ratcliffe, RTSL.eu
Revenue Optimization: How Do You Price?
No matter how well we plan, issues and bugs inevitably occur. Some are
easily solved, but others are far more difficult and complex. This paper
presents a range of largely visual techniques for understanding,
investigating, solving, and monitoring your most difficult problems.
Whether you have an intractable SAS® coding bug or a repeatedly failing
SAS server, this paper offers practical advice and concrete steps to get you
to the bottom of the problem. Tools and techniques discussed in this paper
include Ishikawa (fishbone) diagrams, sequence diagrams, tabular matrices,
and mind maps.
Brenda Carr, Hudsonâs Bay Company Canada
Hudson’s Bay Company, Markdown Optimization – Our Implementation
and Roll-Out Experience, a case study. To further optimize our markdown
spend and benefit from markdown optimization at a style/store level, HBC
upgraded to SAS® Markdown Optimization 4.3. Selling at full price longer
where we can, while still ensuring we achieve our overall seasonal sell
through target allows HBC to fully maximize their markdown spend and
reap the benefits of increased sales and gross margin in better trending
stores. In this session, you will hear about the path HBC took to roll out this
top initiative, how we gained top-down support of the process, and how
we interacted with our business partners to make this tool one that could
be utilized by all areas.
1:30 p.m.
Retailing in the Era of Tech Titans
Lori Schafer, SAS
Paper 387-2013
Over the past decade, five companies have emerged with the potential to
aggressively reshape the landscape of multiple industries – and to change
marketing as we know it. They are the tech titans: Amazon, Apple, eBay,
58
Facebook and Google. Collectively, these companies are worth more than
$1 trillion. Their growth, cash and vision make them formidable
competitors in any industry and complex partners for any company. These
organizations don’t recognize borders – they are marching beyond the
walls of tech into retailing, advertising, publishing, movies, television,
communications, financial services, and eventually into health care and
insurance. The session highlights the strategies of these companies;
retailers may want to consider their own markets and what may happen
because of the tech titans.
3:00 p.m.
Sobeys and SAS - How Do You Talk to Your Customer?
Ashok Setty, Sobeys Inc.
Wanda Shive, SAS
For years, retailers have struggled with measuring the effectiveness of their
promotional advertising efforts. Harnessing the “big data” within their
customer and transaction files continues to be a major challenge.
Approaches for gleaning actionable customer insights from that data are
becoming more common. Measuring total shopping behavior in
conjunction with specific promotion offers provides a better understanding
of the overall impact on profitability. This paper describes how retailers are
utilizing customer analytics to measure the effect that mass promotions
have on the total basket spend of customers and to identify the most
relevant offers for each individual customer.
4:00 p.m.
SAS® Retail Road Map
Saurabh Gupta, SAS
Paper 386-2013
The goal of this presentation is to provide a retail-specific “State of the
Application” update to the SAS® Retail User Group (SRUG) membership. The
session covers the retail solution modules updates released in the past year
and the road map moving forward. This is our forum for the SRUG
membership to hear from the SAS team and vice versa.
5:00 p.m.
Roundtable Discussion: SAS® Integrated Merchandise
Planning
Amy Clouse, Dick's Sporting Goods
This session is designed to be a general discussion with the SRUG
membership on the SAS® Integrated Merchandise Planning Solution. We
will provide an opportunity to ask questions and learn how your peers are
gaining the most value from this SAS® solution.
SAS and Big Data — Room 3001
8:00 a.m.
Leveraging Big Data Using SAS® High-Performance
Analytics Server
Priya Sharma, SAS
Paper 399-2013
With the buzz around big data, see how SAS® High-Performance Analytics
Server enables organizations to create greater business value. This paper
aims at providing best practices and techniques when dealing with big data
analytics. Some of the discussed topics are (1) different methods to load
data for Teradata and Greenplum, (2) checking distribution of loaded data
on all nodes, and (3) new HPDS2 procedure in SAS 9.3. Join this discussion
to learn how SAS High-Performance Analytics Server enables you to use 100
percent of your data to get more precise insights and build complex
models at breakthrough speed, and to see results from an example model.
9:00 a.m.
Big Data Meets Text Mining
Zheng Zhao, SAS
Alicia Bieringer, SAS
James Cox, SAS
Russell Albright, SAS
Paper 400-2013
Learning from your customers and your competitors has become a real
possibility because of the massive amount of Web and social media data
available. However, this abundance of data requires significantly more time
and computer memory to perform analytical tasks. This paper introduces
high-performance text mining techniques for SAS® High-Performance
Analytics. Text parsing, text filtering and dimension reduction are
performed using the new HPTMINE procedure in the SAS High-Performance
Analytics Server and are accessed conveniently from within SAS® Enterprise
Miner™. The paper demonstrates and discusses the advantages of this new
SAS functionality and provides real-world examples of the kinds of
performance improvements you can expect.
10:00 a.m.
Uncovering Patterns in Textual Data with SAS® Visual
Analytics and SAS Text Analytics
Mary Osborne, SAS
Justin Plumley, SAS
Dan Zaratsian, SAS
Paper 403-2013
SAS® Visual Analytics is a powerful tool for exploring big data to uncover
patterns and opportunities hidden in your data. The challenge with big
data is that the majority is unstructured data, in the form of customer
feedback, survey responses, social media conversation, blogs and news
articles. By integrating SAS Visual Analytics with SAS Text Analytics, you can
uncover patterns in big data, while enriching and visualizing your data with
customer sentiment and categorical flags, and uncovering root causes that
primarily exist within unstructured data. This paper highlights a case study
that provides greater insight into big data and demonstrates advanced
visualization, while enhancing time to value by leveraging SAS Visual
Analytics high-performance, in-memory technology, Hadoop, and SAS’
advanced text analytics capabilities.
59
11:00 a.m.
9:30 a.m.
SAS® and Hadoop: The BIG Picture
Using SAS® Enterprise Guide®: A System Administrator’s
Perspective
Paul Kent, SAS
Paper 402-2013
SAS® and Hadoop are made for each other. This talk explains some of the
reasons why they are such a good fit. Examples are drawn from the
customer community to illustrate how SAS is a good addition to your
Hadoop cluster.
SAS® Enterprise Guide® Implementation and
Usage — Room 3002
8:00 a.m.
Consistent and Organized Analysis: Moving Beyond Piein-the-Sky to Actual Implementation via SAS® Enterprise
Guide®
Amber Schmitz, Prime Therapeutics
Paper 404-2013
Consistency and timeliness are two goals that every reporting department
strives to achieve. SAS® Enterprise Guide® provides tools that support these
goals that many analysts overlook. The goal of this paper is to demonstrate
the use of SAS Enterprise Guide to construct a project template that can be
used to report standard metrics for various clients. The template is built
around SAS Enterprise Guide tools that allow for consistent analysis
methods, project organization, and version control documentation. We
explore: 1) Utilizing program code for efficient data pulls, 2) Using SAS
Enterprise Guide tasks for analysis, 3) Exploiting built-in tools such as Notes
and Process Flows for template organization and version control, 4)
Demonstrating business intelligence via built-in graphs.
8:30 a.m.
Haibo Jiang, Allergan, Inc.
Paper 406-2013
This paper shares our experience of supporting SAS® Intelligence Platform
server and client products with other Platform Administrators. We will focus
on SAS® Enterprise Guide® as a client application on Windows desktop, and
SAS servers (SAS Metadata Server®, Object Spawner, and SAS Application
Servers) installed on Hewlett-Packard (HP) UNIX machine. The content of
this paper will describe technical details related to user- and system-related
activities in the following areas: [] Starting SAS servers and checking their
status. [] Connection to SAS servers, user authorization, and authentication
[] Initiation of requests from SAS Enterprise Guide, and using SAS
Workspace Server and Stored Process Server. [] Performance considerations
for the processing and presentation of analysis results in SAS Enterprise
Guide.
10:00 a.m.
Statistical Analyses Using SAS® Enterprise Guide®
Scott Leslie, MedImpact Healthcaree Systems, Inc.
Conducting statistical analyses involves choosing proper methods,
understanding model assumptions and displaying clear results. The latest
releases of SAS® Enterprise Guide® offer conveniences, such as point-andclick wizards and integrated syntax help, to ease the burden on users. This
tutorial demonstrates how to perform statistics quickly and easily using
some handy features of SAS Enterprise Guide. Examples of multiple linear
regression, logistic regression, and survival analysis are covered as well as
some hints on how to navigate SAS Enterprise Guide menus. This tutorial is
intended for SAS® users with beginning to intermediate experience with
the above-mentioned statistics or those with little SAS Enterprise Guide
experience.
The Concepts and Practice of Analysis with SAS®
Enterprise Guide®
11:00 a.m.
Aaron Hill, MDRC
Due in part to its success helping SAS® programmers leverage their
development skills against the challenges of creating analytic solutions in a
new environment, SAS® Enterprise Guide® continues to gain acceptance as
an enterprise solution for reporting and analytic applications. For
organizations to realize maximum benefit from their investment in SAS
Enterprise Guide, subject-matter experts and a new generation of "SAS
naive" analysts also need to be trained in the use of this tool. The current
work provides a framework for this training by explaining the relationship
between SAS Enterprise Guide and traditional SAS programming,
introducing the basic SAS Enterprise Guide concepts necessary to function
in this environment, and presenting examples of common tasks that will
help these users become immediately productive.
60
Destination Known: Programmatically Controlling Your
Output in SAS® Enterprise Guide®
Paper 408-2013
In a SAS® Enterprise Guide® project with multiple reports and graphics, you
can organize output by selectively sending content to different ODS
destinations embedded within the project. For example, within a single
program, you can embed tables in HTML format, graphics in a SAS® report,
and other output in text. The SAS syntax is simple and gives you
programmatic control over all output and destinations. The result: a wellorganized project with all results in their preferred format.
11:30 a.m.
2:30 p.m.
A tour of new features in SAS Enterprise Guide 4.3, 5.1,
and 6.1
For All the Hats You Wear: SAS® Enterprise Guide® Has
Got You Covered
Lina Clover, SAS
Anand Chitale, SAS
I-kong Fu, SAS
Paper 409-2013
As an Enterprise Guide user or administrator, you have probably recently
upgraded to version 4.3 or 5.1 or are looking forward to going to these
versions or the new 6.1 version. In this paper, we will take you on a tour of
new features available to you in each of these releases so that you can be
more productive with your current or upcoming version as well as obtain a
preview of what's coming in the future. We will cover the differences
between the various versions of SAS Enterprise Guide clients and their
compatibility with the corresponding SAS Server versions, and explain how
to upgrade from your current version with the goal of giving you better
decision points for planning your client /server upgrade strategy.
1:30 p.m.
SAS® Enterprise Guide®: More Than a Gift from Outer
Space
Paper 410-2013
SAS® Enterprise Guide® seem alien to you? Let's walk through the many SAS
Enterprise Guide features using some UFO sightings data. During the
presentation, you will learn some basics, how to change the advanced
options, and also explore some newer features of SAS Enterprise Guide.
Whether a SAS® programmer or an experienced SAS Enterprise Guide user,
you will leave with some practical tips and learn what sightings were
reported to the UFO websites.
2:00 p.m.
Using VBA to Debug and Run SAS® Programs
Interactively, Run Batch Jobs, Automate Output, and
Build Applications
FENG LIU, Genworth Financial
Ruiwen Zhang, SAS
Paper 411-2013
SAS® Enterprise Guide® provides an API that lets users automate almost
every aspect of running Enterprise Guide projects or even SAS® programs.
Visual Basic for Applications (VBA) under Excel provides a rich environment
for debugging and running VBA applications. This paper shows how to use
VBA to access the automation API of Enterprise Guide to do sophisticated
tasks or build your own applications. VBA lets you create SAS programs on
the fly, debug and run programs, analyze SAS “lists,” write logs to files, and
examine SAS ODS. You can accomplish tasks like running PROC EXPORT
automatically, which is not feasible through Enterprise Guide’s main
interface. Advanced SAS users can run batch jobs, schedule jobs in parallel,
or use SAS output as input to other applications.
Chris Hemedinger, SAS
Paper 412-2013
Are you new to SAS® and trying to figure out where to begin? Are you a SAS
programmer, already comfortable with code but unsure about new tools?
Are you a statistician seeking to apply your techniques in a new way? Are
you a data manager, just trying to get your data in shape? Perhaps you're a
Jack (or Jill) of all trades trying to manage work in the simplest way
possible. Regardless of your background, SAS® Enterprise Guide® is full of
features and techniques to help you achieve your objective. In this paper,
we show how you can turn SAS Enterprise Guide into your tool to get work
done immediately, without conforming to an entirely new way of working
just to become productive.
3:30 p.m.
Finally, a Tool for Business Users! A Step-By-Step
Practical Approach to Pharma Sales Reporting Using
SAS® Enterprise Guide® 4.3
Airaha Chelvakkanthan Manickam, Cognizant Technology
Solutions
Paper 413-2013
SAS® Enterprise Guide® can be considered the integrated development
environment (IDE) for SAS® users. SAS Enterprise Guide has powerful data
management capabilities, a sophisticated Query Builder, and data
sampling, ranking, transposing, and even creating and editing data
capabilities. This paper presents a real-time case study of reporting Pharma
sales using SAS Enterprise Guide. It includes capturing drug sales dollars
data and performing business transformations, summarization, and
producing summary charts for executive reports and dashboards. The goal
of this paper is to educate SAS users on how all of these actions are easily
performed by a series of simple clicks in SAS Enterprise Guide 4.3.
4:00 p.m.
Stealing the Admin's Thunder: SAS® Macros to Interact
with the UNIX OS from within SAS® Enterprise Guide®
Thomas Kunselman, Southern California Edison
Paper 414-2013
For SAS® users who are unfamiliar with the UNIX environment, simple tasks
like copying, renaming, or changing the permission settings on a file can be
very non-intuitive. Many of these tasks are not even possible through the
SAS® Enterprise Guide® Server List Window. This paper will present several
SAS macros that can be used to: view and kill UNIX host processes; display,
compare, and manage folders and files, including copying subfolders and
changing permissions and owners; display and set default file system
permissions for new objects. Please note that for these macros to work, the
X command must be allowed on the SAS server.
61
4:30 p.m.
Improving Your Relationship with SAS® Enterprise
Guide: Tips from SAS® Technical Support
9:30 a.m.
Jennifer Bjurstrom, SAS
Paper 415-2013
SAS® Enterprise Guide® has proven to be a very beneficial tool for both
novice and experienced SAS® users. Because it is such a powerful tool, SAS
Enterprise Guide has risen in popularity over the years. As a result, SAS
Technical Support consultants field many calls from users who want to
know the best way to use the application to accomplish a task or to obtain
the results they want. This paper encompasses many of the tips that SAS
Technical Support has provided to customers over the years. These tips are
designed to improve your proficiency with SAS Enterprise Guide in the
areas of connection and configuration, workflow preferences, logging, data
manipulation, project files, X commands, and custom tasks.
8:00 a.m.
Finding the Gold in Your Data: An Overview of Data
Mining
David Dickey, NC State University
"Data mining" has appeared often recently in analytic literature and even in
popular literature, so what exactly is data mining and what does SAS®
provide in terms of data mining capabilities? The answer is that data mining
is a collection of tools designed to discover useful structure in large data
sets. With an emphasis on examples, this talk gives an overview of methods
available in SAS® Enterprise Miner™ and should be accessible to a general
audience. Topics include predictive modeling, decision trees, association
analysis, incorporation of profits, and neural networks. We'll see that some
basic ideas underlying these techniques are related to standard statistical
techniques that have been around for some time but now have been
automated to become more user friendly.
9:30 a.m.
Current Directions in SAS/STAT® Software Development
Maura Stokes, SAS
Paper 432-2013
Recent years have brought you SAS/STAT® releases in rapid succession, and
another one is coming in 2013. Which new software features will make a
difference in your work? What new statistical trends should you know
about? This paper describes recent areas of development focus, such as
Bayesian analysis, missing data methods, postfitting inference, quantile
modeling, finite mixture models, specialized survival analysis, structural
equation modeling, and spatial statistics. The paper introduces you to the
concepts and illustrates them with practical applications.
From Big Data to Big Statistics
John Sall, SAS
Paper 434-2013
Now that we have lots of data and can process it amazingly fast, we still
need ways to look at it without being overwhelmed. We don’t want to look
at 10,000 graphs--we want one graph that shows the bright spots among
10,000 graphs. We need volcano plots and false-discovery-rate plots. We
want the computer and software to do the work of finding what is most
interesting and bringing it to our attention. We want our results sorted and
summarized, but with access to the detail we need to understand it. Also,
when we look at the most significant of thousands of statistical tests, we
want to know if we are seeing random coincidence selected out of
thousands, or if we are seeing real effects.
10:30 a.m.
DICHOTOMIZED_D: A SAS® Macro for Computing Effect
Sizes for Artificially Dichotomized Variables
Patrice Rasmussen, 5336 Clover Mist Drive
Isaac Li, Univ. of South Florida
Jeanine Romano, University of South Florida
Aarti Bellara, University of South Florida
Harold Holmes, University of South Florida
Paper 491-2013
Measures of effect size are recommended to communicate information
about the strength of relationships between variables, providing
information to supplement the reject/fail-to-reject decision obtained in
statistical hypothesis testing. With artificially dichotomized response
variables, seven methods have been proposed to estimate the standardized
mean difference effect size that would have been realized before
dichotomization. This paper provides a SAS® macro, DICHOTOMIZED_D, for
computing these seven effect size estimates by utilizing data from FREQ
procedure output data sets. The paper provides the macro programming
language, as well as results from an executed example of the macro.
10:30 a.m.
A Multilevel Model Primer Using SAS® PROC MIXED
Bethany Bell, University of South Carolina
Mihaela Ene, University of South Carolina
Whitney Smiley, University of South Carolina
Jason Schoeneberger, University of South Carolina
Paper 433-2013
This paper provides an introduction to specifying multilevel models using
PROC MIXED. After a brief introduction to the field of multilevel modeling,
users are provided with concrete examples of how PROC MIXED can be
used to estimate (a) two-level organizational models, (b) two-level growth
62
models, (c) three-level organizational models, and (4) three-level growth
models. Both random intercept and random intercept and slope models are
illustrated. Examples are shown using Early Childhood Longitudinal Study–
Kindergarten cohort data. For each example, narrative explanations are
accompanied by annotated examples of the PROC MIXED code and
corresponding output. Users are also introduced to examining model fit
using the SAS® macro MIXED_FIT as well as checking the distributional
assumptions for two-level models using the SAS macro MIXED_DX.
11:00 a.m.
The Value of Neighborhood Information in Prospect
Selection Models: Investigating the Optimal Level of
Granularity
Philippe Baecke, Vlerick Business School
Dirk Van den Poel, Ghent University
Within analytical customer relationship management (CRM), customer
acquisition models suffer the most from a lack of data quality because the
information of potential customers is mostly limited to socio-demographic
and lifestyle variables obtained from external data vendors. Particularly in
this situation, taking advantage of the spatial correlation between
customers can improve the predictive performance of these models. This
study compares the predictive performance of an autoregressive and
hierarchical technique in an application that identifies potential new
customers for 25 products and brands. In addition, this study shows that
the predictive improvement can vary significantly depending on the
granularity level on which the neighborhoods are composed. Therefore, a
model is introduced that simultaneously incorporates multiple levels of
granularity resulting in even more accurate predictions.
1:30 p.m.
Having an EFFECT: More General Linear Modeling and
Analysis with the New EFFECT Statement in SAS/STAT®
Software
Phil Gibbs, SAS
Randy Tobias, SAS
Kathleen Kiernan, SAS
Jill Tao, SAS
Paper 437-2013
Linear models relate a response to a linear function of a design matrix X.
General linear models, long available in standard SAS/STAT® 9.3 procedures
such as GLM and MIXED, incorporate classification, interaction, and
crossproduct effects to define X. The new EFFECT statement, which is
available in many SAS/STAT 9.3 procedures, extends how you can define X.
It enables you to fit models with nonparametric regression effects,
crossover and carryover effects, and complicated inheritance effects. This
paper first shows how the EFFECT statement fits into the general
architecture of SAS/STAT linear modeling tools, and then explains and
demonstrates specific effect types. You will see how this powerful new
feature easily enhances the statistical analyses that you can perform.
1:30 p.m.
Missing No More: Using the MCMC Procedure to Model
Missing Data
Fang Chen, SAS
Paper 436-2013
11:00 a.m.
Joint Modeling of Mixed Outcomes in Health Services
Research
Joseph Gardiner, Michigan State University
Outcomes with different attributes, of continuous, count, and categorical
types, are often encountered jointly in many settings. For example, two
widely used measures of healthcare utilization, length of stay (LOS) and
cost, can be analyzed jointly with LOS as a count and cost as continuous.
Occurrence of an adverse event (binary) would impact both outcomes. For
fitting marginal distributions and assessing the impact of explanatory
variables on outcome, SAS offers a number of procedures. Correlation and
clustering are additional features of these outcomes that must be
addressed in analyses. This paper surveys the GLIMMIX, COPULA, PHREG,
and QLIM procedures, which can be applied to modeling multivariate
outcomes of mixed types. Examples from the literature are used to
demonstrate the application of these procedures.
Missing data are often a problem in statistical modeling. The Bayesian
paradigm offers a natural model-based solution for this problem by
treating missing values as random variables and estimating their posterior
distributions. This paper reviews the Bayesian approach and describes how
the MCMC procedure implements it. Beginning with SAS/STAT® 12.1, PROC
MCMC automatically samples all missing values and incorporates them in
the Markov chain for the parameters. You can use PROC MCMC to handle
various types of missing data, including data that are missing at random
(MAR) and missing not at random (MNAR). PROC MCMC can also perform
joint modeling of missing responses and covariates.
2:30 p.m.
Regression of NASCAR: Looking into Five Years of
Jimmie Johnson
Yun Gao, California State Universiday Long Beach
Paper 439-2013
In this paper, we investigate the winnings associated with different factors
for NASCAR drivers. We want to predict the winnings that a driver can earn
in a season given other, related factors, such as the number of races the
driver competes in, the average finish position, or the make of car. We
obtained 190 observations with 15 factors and randomly split the data into
learning data and test data. Using the learning data set, we conducted
multiple regression analyses to build a predictive model. Then we
63
examined the final model with the test data set to see how well the model
would work in the future. The model shows a high degree of accuracy in
predicting the future.
variance estimates. Likelihood ratio testing is a more flexible approach, as it
can be used to compare models that differ in both fixed and random
effects. The likelihood ratio test statistic requires a complex calculation that
is not included in PROC MIANALYZE. This paper describes a SAS macro,
MMI_ANALYZE, that fits two user-specified models in PROC MIXED, pools
the estimates from those models (including variance components), and
implements a pooled likelihood ratio test.
2:30 p.m.
A SAS® Macro for Applying Multiple Imputation to
Multilevel Data
Paper 438-2013
Multilevel Reweighted Regression Models to Estimate
County-Level Racial Health Disparities Using PROC
GLIMMIX
Stephen Mistler, Arizona State University
Single-level multiple imputation procedures (e.g., PROC MI) are not
appropriate for multilevel data sets where observations are nested within
clusters. Analyzing multilevel data imputed with a single-level procedure
yields variance estimates that are biased toward zero and may yield other
biased parameters. Given the prevalence of clustered data (e.g., children
within schools; employees within companies; observations within people),
a general approach is needed for handling missing data in multilevel data
sets. This paper describes a SAS® macro, MMI_IMPUTE, that performs
multiple imputation for clustered data sets with two levels. The macro uses
a Bayesian implementation of the mixed linear model to generate
imputations for lower-level incomplete variables, and uses single-level
procedures similar to those used in PROC MI to generate imputations for
cluster-level variables.
3:00 p.m.
Short-Term Costs of Smoking during Pregnancy: A
Geometric Multidimensional Approach
3:30 p.m.
Melody S. Goodman, Division of Public Health Sciences at
Washington University in St. Louis School of Medicine
Lucy D'Agostino, Division of Public Health Sciences,
Department of Surgery, Washington University School of
Medicine
Paper 442-2013
The agenda to reduce racial health disparities has been set primarily at the
national and state levels. These levels may be too far removed from the
individual level where health outcomes are realized, and this disconnect
may be slowing the progress made in reducing these disparities. This paper
focuses on establishing county-level prevalence estimates of diabetes
among non-Hispanic whites and non-Hispanic blacks. These estimates use
multilevel reweighted regression models through the GLIMMIX procedure
with 2010 Behavioral Risk Factor Surveillance System data and 2010 census
data. To examine whether racial disparities exist at the county level, the
paper estimates the risk difference of prevalence estimates between races.
It subsequently ranks counties and states by the magnitude of disparities.
Violeta Balinskaite, University of Bologna
Paper 441-2013
Smoking during pregnancy imposes a considerable economic burden on
society. This phenomenon has been studied fairly extensively in the United
States, but little is known about its costs within the European Union. This
paper attempts to estimate the additional neonatal costs of a mother in the
European Union who smokes during pregnancy compared to the
alternative of her not smoking. The geometric multidimensional approach
that is used for analysis involves the use of conditional multiple
correspondence analysis as a tool for investigating the dependence
relationship between covariates and the assignment-to-treatment indicator
variable within a strategy whose final aim is to find balanced groups.
3:00 p.m.
A SAS® Macro for Computing Pooled Likelihood Ratio
Tests with Multiply Imputed Data
Stephen Mistler, Arizona State University
Paper 440-2013
For multilevel analyses (e.g., linear mixed models), researchers are often
interested in pooling, interpreting, and testing both fixed effects and
random effects. PROC MIANALYZE has two shortcomings in this regard.
First, it cannot easily pool variance estimates. Second, the significance tests
of these estimates are Wald-type tests that are inappropriate for testing
64
3:30 p.m.
Commuting Time and Accessibility in a Joint Residential
Location, Workplace, and Job Type Choice Model
Ignacio A. Inoa, Université de Cergy-Pontoise
Paper 443-2013
The effect of an individual-specific measure of accessibility to jobs is
analyzed using a three-level nested logit model of residential location,
workplace, and job-type choice. This measure takes into account the
attractiveness of different job types when the workplace choice is
anticipated in the residential location decision. The model allows for
variation in the preferences for job types across individuals and accounts
for individual heterogeneity of preferences at each choice level in the
following dimensions: education, age, gender, and children. Using data
from the Greater Paris Area, estimation results indicate that the individualspecific accessibility measure is an important determinant of the residential
location choice and its effects strongly differ along the life cycle.
4:00 p.m.
Models for Ordinal Response Data
Robin High, University of Nebraska Medical Center
Paper 445-2013
The types of computations with response data having ordered categories
with SAS® procedures are not well known. Various models can be evaluated
through programming statements entered into PROC NLMIXED including
the partial proportional odds, adjacent logit, continuation ratio, and
stereotype models. The process requires no restructuring of the input data
set, as required with procedures that can produce a few of these models.
The correct interpretation of ordinal logistic regression models depends on
how both the response and explanatory data are coded and if any formats
are applied. Implementation of these models assumes a background with
general linear models and categorical data analysis including maximum
likelihood equations and computing odds ratios with binary data.
4:00 p.m.
Examining Mediator and Indirect Effects of Loneliness in
Social Support on Social Well-Being Using the Baron and
Kenny Method and a Bootstrapping Method
Abbas Tavakoli, University of South Carolina
Sue Heiney, University of South Carolina
Paper 444-2013
This study examines the mediator effect and the indirect effect of loneliness
in social support on social well-being by using two methods: the Baron and
Kenny method and a bootstrapping method. The cross-sectional data come
from a longitudinal randomized trial design that had 185 participants.
Baron and Kenny steps and Hayes were used to examine the mediator
effect. The Baron and Kenny results indicate no mediator effect for
loneliness in the relationship between social support and social well-being.
Bootstrapping results indicate that the direct effect was 0.591 (95% CI:
0.589-0.593 for normal theory and 0.481- 0.690 for percentile) and the
indirect effect was 0.040 (95% CI: 0.039-0.040 for normal theory and
0.006-0.087 for percentile). The results show that both methods have
significant indirect effect.
4:30 p.m.
Ordinal Response Modeling with the LOGISTIC
Procedure
responses. This paper also discusses methods of determining which
covariates have proportional odds. The reader is assumed to be familiar
with using PROC LOGISTIC for binary logistic regression.
4:30 p.m.
Segmentation and Classification Analysis Using SAS®
Rachel Poulsen, TiVo
An idiom in the customer service industry is “the customer is always right”.
However, in many instances the customer will not speak up and another
popular idiom must be used “Actions speak louder than words”. Customer
actions can be measured to infer what they will not say. Once measured,
segmentation analysis can be used to make sense of the large amount of
behavioral data by placing customers into various segments. Classification
models are then used to assign new customers to a segment. Statistical
algorithms used to segment and classify observations include Collaborative
Filtering and Machine Learning Models. This paper will illustrate how SAS®
can be used to segment and classify observations using the FASTCLUS and
DISCRIM procedures.
2006
8:00 a.m.
Enhance Your High Availability Story by Clustering Your
SAS® Metadata Server in SAS® 9.4
Bryan Wolfe, SAS
Amy Peters, SAS
Paper 468-2013
It can be challenging to use clustered file systems, backups and system
tools to ensure SAS® Metadata Server is always available. In SAS 9.4, you can
create and manage a clustered metadata deployment using SAS tools to
remove single-point-of-failure concerns for the SAS Metadata Server. The
clustered SAS Metadata Server keeps its data in sync, balances its load and
continues to handle requests if a node fails, all while presenting a single
face to the outside world. Once it is set up, you don’t need to treat the SAS
Metadata Server any differently than you do the single server.
8:30 a.m.
Best Practices for Deploying Your SAS® Applications in a
High-Availability Cluster
Helen Pan, SAS
Bob Derr, SAS
Paper 469-2013
Logistic regression is most often used to model simple binary response
data. Two modifications extend it to ordinal responses that have more than
two levels: using multiple response functions to model the ordered
behavior, and considering whether covariates have common slopes across
response functions. This paper describes how you can use the LOGISTIC
procedure to model ordinal responses. Before SAS/STAT® 12.1, you could
use cumulative logit response functions with proportional odds. In SAS/
STAT 12.1, you can fit partial proportional odds models to ordinal
Are you frustrated when software or hardware failures interrupt your SAS®
system? A high-availability (HA) cluster with HA software can help you by
providing failover protection for your SAS applications, thus reducing your
system downtime. This paper discusses what you should consider when
deploying your SAS applications or servers in an HA cluster, as well as the
following best practices of the deployment process: - HA cluster
architecture with a SAS deployment. - SAS installation and deployment in
an HA cluster. - Dependency configuration of the SAS servers. - Impact on
the SAS clients when a failover happens.
Paper 446-2013
65
9:00 a.m.
11:00 a.m.
The Top Four User-Requested Grid Features Delivered
with SAS® Grid Manager 9.4
Best Practices for Deploying SAS® on Red Hat Enterprise
Linux
Paper 470-2013
The number of SAS deployments on Red Hat Enterprise Linux (RHEL)
continues to increase in recent years because more and more customers
have found RHEL to be the best price/performance choice for new and/or
updated SAS deployments on x86 systems. Back for the third year at SGF,
Shak and Barry will share new performance findings and best practices for
deploying SAS on Red Hat Enterprise Linux and will discuss topics such as
virtualization, GFS2 shared file system, SAS Grid Manager and more. This
session will be beneficial for SAS customers interested in deploying on Red
Hat Enterprise Linux, or existing SAS-on-RHEL customers who want to get
more out of their deployments.
Doug Haigh, SAS
As more and more SAS customers choose to deploy their platform for SAS®
Business Analytics with SAS Grid Manager, we continue to expand the
capabilities of a shared, managed, and highly available environment.
Several features have been added in SAS 9.4, providing the following: easier administration of the different types of SAS users in the grid increased management of more SAS components running in the grid integration of the grid with IT standards, such as an existing enterprise
scheduler - increased debugging capabilities for quick problem resolution
This paper details how to accomplish all of the above through new grid
options sets, workspace servers spawned on the grid, new options in
SASGSUB, and enhanced error logging.
9:30 a.m.
Creating metadata environment from existing one for
testing purpose
Jouni Javanainen, Aureolis
Paper 471-2013
Most organizations have a need for the development, test and production
environments, either in the same physical platform or on separate
platforms. Separate environments may not use the same ports of
communication. Using package of migration from existing metadata, it is
possible to define specific communication ports, so that it does not disturb
the other environments. In addition to this it is needed very little extra
finishing steps, such as paths of directories or libraries. It is important that
this type of environment can be created through a formal process quickly,
reliably and efficiently. SAS has a good set of tools to create this a welldocumented method for creating environments. This paper covers how
easily you can create identical environments for development and testing
purposes.
10:00 a.m.
Configure Your Foundation SAS® Client Easily and
without Risk
Peter Crawford, Crawford Software Consultancy Limited
For your client install of Base SAS®, don't use the default provided. Instead,
use the simple features in this paper and presentation to support the
flexibility you want. This method of applying options as SAS starts,
eliminates risky techniques of the past when developers would update the
configuration file provided by SAS. The technique is described and
demonstrated with examples for the Microsoft Windows environment for
Base SAS, but the issues are very similar if launching SAS clients and servers
on UNIX (including UNIX on z/OS). There is a small overhead with this
proposal, but if you review the paper, I’ m sure you will consider it worthy
enough to give it a try.
66
1:30 p.m.
SAS Administration: The State of the Art (Panel
Discussion)
Greg Nelson, ThotWave Technologies, LLC.
Paul Homes, Metacoda
Jennifer Parks, CSC Inc
The implementation of SAS can take on many forms in organizations
around the globe - from single-server SAS Foundation installs to multi-tenet
SAS solutions. Similarly the role of the SAS administrator has evolved
significantly - especially since the introduction of SAS 9. Join us for this
panel discussion for an in-depth conversation on SAS Administration.
Topics covered will include:
• SAS administrator roles and responsibilities
• Best practices in data management, governance, architecture and
business process integration
• Adoption of new technologies including upgrades and maintenance
• Backup, Recovery, Disaster Recovery
• Multi-tenet architectures
• Optimizing SAS. Panelists will include experts in capacity planning, SAS
metadata, security, optimizing operating environments, system design
and architecture. Panelists will include personnel from SAS, system
integrators and various industry organizations
2:30 p.m.
A Case Study of Tuning an Enterprise Business
Intelligence Application in a Multi-OS Environment
Fred Forst, SAS
Paper 474-2013
A corporation is planning to expand its use of SAS® Web Report Studio by
increasing its user base. Two questions emerge: 1) What happens to SAS
Web Report Studio response time as users are added? 2) What tuning
modifications can be used to improve performance? This case study uses a
SAS Web Report Studio simulation technique to expand the workload and
examine software response times. Logs from SAS Web Report Studio,
WebSphere, NMON and the server tier are parsed and stored in a SAS data
set for all analyses. Many stress test runs are executed, modifying various
tuning parameters. The analysis shows the effects of tuning and offers
insight on best practices.
3:00 p.m.
Grand Designs: Why It Pays to Think About Technical
Architecture Design Before You Act
Simon Williams, SAS
Paper 475-2013
The sustainability of a business is not a short-term goal; it depends on
effective planning and considered thinking. The same principle of
sustainability applies to IT systems, including those using technologies
from SAS. Getting the best out of SAS® technology is easy once all the IT
and business requirements and constraints are clearly defined, understood
and agreed upon as being testable. This paper describes the effective
thinking and systematic approach that can be applied to creating a
sustainable SAS solution. The paper also outlines the advantages to this
approach and describes a simple checklist that can be helpful when
designing a new SAS environment.
3:30 p.m.
Kerberos and SAS® 9.4: A Three-Headed Solution for
Authentication
Stuart Rogers, SAS
Paper 476-2013
Kerberos is a network authentication protocol designed to provide strong
authentication for client/server applications by using secret-key
cryptography. With the release of SAS® 9.4, there are three ways Kerberos
can be used with the SAS platform for business analytics. Kerberos provides
Integrated Windows authentication from a range of clients to a range of
servers. This paper reviews how Kerberos is used with the SAS 9.4 platform
for business analytics. It explores the considerations and constraints when
using Kerberos and summarizes solutions for some common issues.
investigated and implementation client virtualization to simplify the
configuration, deployment and support of their end-user computers. Client
virtualization looks at changing the 1 end-user to 1 desktop computer
paradigm for SAS software installation and finding ways of reducing the
administrative burden associated with the one end user desktop computer
while gaining operational efficiencies and a more robust deployment
model. This paper will focus on the following topics: Primary drivers for
adopting this technology; Census SAS support model; Client virtualization
architecture; Deployment best practices.
5:00 p.m.
SAS® Enterprise Business Intelligence Deployment
Projects in the Federal Sector: Best Practices
Jennifer Parks, CSC Inc
Systems engineering life cycles (SELC) in the federal sector embody a high
level of complexity due to legislative mandates, agency policies, and
contract specifications layered over industry best practices, all of which
must be taken into consideration when designing and deploying a system
release. Additional complexity stems from the unique nature of ad-hoc
predictive analytic systems that are at odds with traditional, unidirectional
federal production software deployments to which many federal sector
project managers have grown accustomed. This paper offers a high-level
roadmap for successful SAS® EBI design and deployment projects within
the federal sector. It's addressed primarily to project managers and SAS
administrators engaged in the SELC process for a SAS EBI system release.
4:00 p.m.
SAS® and the New Virtual Storage Systems
Tony Brown, SAS
Margaret Crevar, SAS
Paper 487-2013
Storage providers are offering a wave of new, advanced storage
subsystems. These offerings promise virtualized, thin-provisioned, tiered,
intelligent storage that is easy to manage and will reduce costs. For many
random and mixed workload applications, the promises deliver well and
with good performance. Unfortunately, the SAS® I/O workload profiles tend
to violate some of the primary design assumptions underlying the
configuration of these new systems. This paper will address the SAS
workload-specific issues that need to be considered when configuring
these new storage systems for expected performance.
4:30 p.m.
SAS® Virtual Desktop Deployment at the U.S. Bureau of
the Census
Lori Guido, US Census Bureau
Michael Bretz, SAS
Stephen Moore, US Census Bureau
Paper 490-2013
The U.S. Census Bureau has a SAS® user base of approximately 2600 users
requiring deployment of many SAS client solutions on individual desktops.
Using new deployment strategies, we reduced deployment delivery time
while increasing installation quality and standardization. Census
67
68
11:30 a.m.
9:00 a.m.
A First Look at the ODS Destination for PowerPoint
Creating Graph Collections with Consistent Colors Using
ODS Graphics?
Paper 041-2013
Philip Holland, Holland Numerics Ltd
Collections of line graphs or bar charts, where the graph data is grouped by
the same value, are frequently used to identify differences and similarities
in behavior. Unfortunately, by default, the colors used for each line can
change across the graph collection if some group values are not present in
every graph. In SAS/GRAPH®, this problem has been solved by generating
SYMBOL or PATTERN statements based on the data, or using annotation to
create all of the graph lines, bars and legends. Neither of these solutions is
readily available in ODS Graphics. This paper will solve this problem using
macros with PROC SGPLOT and PROC TEMPLATE, giving the user complete
control over how every graph looks.
10:00 a.m.
Renovating Your SAS® 9.3 ODS Output: Tools for
Everything from Minor Remodeling to Extreme
Makeovers
Bari Lawhorn, SAS
Paper 039-2013
The SAS® 9.3 HTML output that is generated with the ODS HTML statement
and the new HTMLBLUE style looks great with the default settings. But
realistically, there are always pieces of your output that you want to
change. Sometimes you just need a little remodeling to get the results you
want; other times, the desired changes call for an extreme makeover. This
paper shows how you can use certain ODS statements (for example, ODS
SELECT and ODS TEXT) and ODS statement options (for example, STYLE=)
for minor remodeling. The paper also illustrates how to use the REGISTRY,
TEMPLATE, and DOCUMENT procedures for a more extreme makeover.
These makeovers apply to HTML as well as other ODS destinations.
10:30 a.m.
"How Do I ...?" There Is More Than One Way to Solve That
Problem; Why Continuing to Learn Is So Important
In the SAS® forums, questions are often posted that start with "How do
I . . . ?". Generally, there are multiple solutions to the posted problem, and
these vary from simple to complex. All too often, the simple solution is both
inefficient and reflects a naive understanding of the SAS language. This
would not be so very bad except sometimes the responder thinks that their
response is the best solution or, perhaps worst, the only solution.
Tim Hunter, SAS
The inclusion of PowerPoint is part of the next generation of ODS
destinations. You can use this destination to send PROC OUTPUT directly
into native PowerPoint format. See examples of slides created by ODS.
Learn how to create presentations using ODS; how to use ODS style
templates to customize the look of your presentations; and how to use
predefined layouts to make title slides and two-column slides. Learn how
the ODS destination for PowerPoint is similar – and different – to other ODS
destinations. Stop cutting and pasting; let the ODS destination for
PowerPoint do the work for you!
12:30 p.m.
Can You Create Another PowerPoint for Me? How to Use
Base SAS® and DDE to Automate Snappy PowerPoint
Presentations
Scott Koval, Pinnacle Solutions, Inc.
Mitchell Weiss, Maguire Associates
Paper 042-2013
Your supervisor appreciates your wonderful and informative SAS® reports.
How many times have you heard, “Great! Now, can you compile all the SAS
reports into a PowerPoint presentation?” At that moment, you wish you
could press a button to automate the process because SAS programmers
spend way too much time updating PowerPoint slides. This paper offers
solutions to make your life easier by building upon techniques from Koen
Vyverman’s paper (SUGI 30, 2005) that discussed the Dynamic Data
Exchange (DDE) feature within SAS to write through MS Excel to MS
PowerPoint. The goal is to free data analysts from PowerPoint tyranny by
enabling efficient and repeatable PowerPoint presentations.
9:00 a.m.
How Am I Driving - My Business? (Techniques, from the
Insurance Industry That Can Be Applied to Other
Business Areas to "Drive" Better Performance)
Guy Garrett, Achieve Intelligence
Steve Morton, Applied System Knowledge Ltd
This paper runs through the high level strategic measurements that general
insurance companies need to routinely monitor, looking at the technical
solutions available today using SAS® software. The techniques used in this
paper can also be applied to other industries helping executives and
managers to measure and monitor their businesses’ performance.
69
9:00 a.m.
12:00 p.m.
Emerging Best Practices in the Age of Democratized
Analytics
Popular Tips and Tricks to Help You Use SAS® Web
Report Studio More Efficiently
Paper 541-2013
Paper 062-2013
Welcome to the age of analytics, everyone! Fact-based decision making has
never been more pervasive than today. Reports, dashboards and mobile BI
applications can empower entire organizations to understand and act on
analytics in the office and on the go. Your organization, team or division is
curious about data exploration capabilities and yet there are very diverse
levels of analytics understanding from person to person. Attend this
presentation for an overview of the best practices you can use to satiate the
curiosity of business decision makers to explore data, while maintaining
analytic integrity of your models and processes.
For more than six years, SAS® Web Report Studio has enabled users at all
skill levels to create, view, and explore centrally stored reports. This paper
discusses tips and tricks for reporting techniques that have been most
popular with customers over the years. The paper also explains new
features that shipped with the second maintenance release of SAS Web
Report Studio 4.31. As with the tips and tricks, these new features offer
more efficient methods for tasks related to conditional highlighting and to
content enhancement in reports that are sent via e-mail. The techniques
and features that are discussed cover tasks in the following key areas:
performance, filtering, scheduling, and distribution, report design, and
sending reports via e-mail.
Justin Choy, SAS
Keith Myers, SAS
10:00 a.m.
How Mobile Changes the BI Experience
Murali Nori, SAS
Paper 053-2013
The advent of a new generation of tablets catapulted corporations’ use of
mobile devices. With SAS® Mobile BI for tablets, anyone who uses BI for
work and decision making has a new way to experience BI content. This
paper presents some end-to-end use cases to demonstrate how
revolutionary the user experience is with SAS Mobile BI. It also
demonstrates how easy it is to access and navigate BI content. Discover
how BI on mobile devices changes the user experience and the reach of BI
content for productivity, decision making and extracting better ROI.
11:00 a.m.
What’s New in SAS® Enterprise Business Intelligence for
SAS® 9.3
Rick Styll, SAS
Paper 060-2013
SAS® Enterprise BI Server provides a comprehensive suite of BI tools that
allows a broad set of business and IT users to produce and consume
consistent, fact-based information. The latest revision contains
enhancements to both SAS Web Report Studio and SAS BI Dashboard. Key
capabilities are discussed and demonstrated by members of the product
team. Designing reports and dashboards is now more flexible, and
downstream consumers benefit from better performance, improved
navigation and interactions, and better integration with Excel and your
email client. Plans for future releases are previewed, such as mobile delivery
of SAS Web Report Studio reports, and how SAS Enterprise BI Server fits
within the overall BI portfolio.
70
Customer Intelligence — Room 2001
9:00 a.m.
SAS® Treatments: One to One Marketing with
Customized Treatment Processes
Dave Gribbin, SAS
Amy Glassman, SAS
Paper 065-2013
The importance of sending the right message and offer to the right person
at the right time has never been more relevant than in today’s cluttered
marketing environment. SAS® Marketing Automation easily handles
segment-level messaging with out-of-the-box functionality. But how do
you send the right message and the appropriately valued offers to the right
person? And how can an organization efficiently manage many treatment
versions across distinct campaigns? This paper presents a case study of a
casino company that sends highly personalized communications and offers
to its clientele using SAS Marketing Automation. It describes how
treatments are applied at both a segment and one-to-one level. It outlines a
simple custom process that streamlines versioning treatments for reuse in
multiple campaigns.
10:00 a.m.
You’re Invited! Learn How SAS Uses SAS® Software to
Invite You to SAS® Global Forum
Lori Jordan, SAS
Shawn Skillman, SAS
Paper 066-2013
Discover how you are chosen to receive an invitation to SAS® Global Forum.
This paper explores the use of SAS® software, including SAS® Marketing
Automation, SAS® Enterprise Guide®, SAS® Enterprise BI, and SAS® DataFlux,
in the selection process. See how SAS Marketing uses strategic
segmentation practices and advanced analytics to target email
communications and improve list performance for SAS Global Forum.
11:00 a.m.
Financial Services — Room 2010
Hot off the Press: SAS® Marketing Automation 6.1
9:30 a.m.
Mark Brown, SAS
Brian Chick, SAS
Paper 067-2013
SAS® Marketing Automation 6.1 is a major release that introduces a new
browser-based user interface for business analysts. The new look and feel
deliver a highly interactive and intuitive user experience while delivering
key customer-driven features, including support for control groups, live
seeds, enhanced campaign definitions, and improved scheduling and
execution. This paper introduces new navigation, improved search
capabilities and easier sharing of information; it also presents easier reuse
of treatments, campaign components, scheduling and control group
methodologies. Various control group techniques will be presented,
including: - Hold-out control groups. - A/B testing control groups. Champion/challenger. - Challenger/challenger. Business case studies based
on customer examples illustrate the matching of campaign business goals
with the appropriate control group technique and why they are needed for
robust marketing measurement.
12:00 p.m.
Product Affinity Segmentation That Uses the Doughnut
Clustering Approach
Darius Baer, SAS
Paper 068-2013
Product affinity means the natural liking of customers for products. Product
affinity segmentation divides customers into groups based on purchased
products. While conceptually appealing to marketers and business analysts,
in practice it often yields inappropriate solutions such as one large segment
and many tiny segments. Standard transformations such as logarithms do
not help. In this paper, we demonstrate how a combination of softmax
transformations with a doughnut clustering approach (single central
cluster) results in more evenly sized product affinity segments for 30,000
customers of a business-to-business company. The affinity segments show
meaningful differences in product buying patterns across the customer
base, and can be used for identifying cross-selling and up-selling
opportunities. The segments are further profiled using customers'
background variables to provide deeper business insights.
12:30 p.m.
Predicting Women's Department Purchases in a Retail
Store By Using the SEMMA Methodology
Michael Soto, Ripley
Paper 069-2013
One of our focus areas is to improve the business in the Women's
Department because it is currently our most powerful department in terms
of transactions originated by customers. It raises the need to implement an
analytical model focused on this department for establishing what offer is
the most appropriate for our customers according to buying patterns of
customers, augmenting the likelihood that a customer comes back to the
stores. Those patterns are calculated based on demographic transactional
data, and any other interaction that our customers have had with our
stores. The predictive model we used is the logistic regression, and it was
executed following the SEMMA methodology considered by SAS® for
projects in SAS® Enterprise Miner™.
Managing and Analyzing Financial Risk on Big Data with
High-Performance Risk and Visual Analytics
Cary Orange, SAS
Donald Erdman, SAS
Stacey Christian, SAS
Paper 110-2013
SAS® High-Performance Risk is a distributed forecasting engine for financial
risk, used to compute such things as value at risk (VaR). The output created
by this engine can be voluminous and unwieldy since it represents future
portfolio prices, which can be billions of rows long. It is desirable to postprocess these results to produce ad hoc reports. For performance, it is
important to keep these results in parallel. This is the perfect situation to
use high-performance, in-memory solutions. In this paper, we present
examples and results of running SAS High-Performance Risk scenarios and
analyzing them with SAS Visual Analytics.
10:30 a.m.
A Case Study in Firmwide Stress Testing: Engineering
the CCAR Process
Carsten Heiliger, Sun Trust
Stress testing has become pervasive. The trouble lies in isolating the
substantive, insightful activity from the overwrought chaff. Almost an
overused platitude, the term “stress testing” can be inserted into just about
any process in a financial institution, and there will be an army of
consultants claiming to have a best-practice opinion on the topic. The
reality is far more convoluted. Often, what is referred to as a stress test is
simply a sensitivity analysis with a focus on a suboptimal outcome. Other
times, an operationally focused risk assessment is termed a stress test, as it
is analyzing processes performing below an optimal level.
11:30 a.m.
Integrated Framework for Stress Testing in SAS®
Jimmy Skoglund, SAS
Wei Chen, SAS
Paper 112-2013
Stress testing is an integrated part of enterprise risk management and is a
regulatory requirement. Stress testing is especially useful for integrating
forward-looking views into risk analysis. Indeed, stress tests can provide
useful information about a firm’s risk exposure that statistical risk methods,
calibrated on the basis of history, can miss. However, traditional stress
testing is done on a stand-alone basis. This makes the interpretation of risk
obtained from stress events vs. from risk analysis with statistical models
difficult to interpret. We consider a Markov model and innovative
implementation in SAS® that integrates rare stress events into regular
statistical risk models. The model allows a consistent integration of the
information in backward-looking historical data.
71
12:30 p.m.
11:00 a.m.
Detecting Cross-Channel Fraud Using SAS®
Creating Clark Error Grid with SAS/GRAPH®, the SAS/
GRAPH Annotate Facility, and SAS® Macro Applications
Krutharth Kumar Peravalli Venkata Naga, Oklahoma State
University
Yongyin Wang, Medtronic Diabetes
John Shin, Medtronic Diabetes
Paper 113-2013
Paper 133-2013
In a world where criminals are getting effective in their ability to gain
information about a customer of a particular bank, cross-channel
monitoring and assessment has become very important. As each day
passes by, criminals are also getting bolder in terms of engaging beyond a
single channel to set in motion the movement of money. In these scenarios,
a cross-channel review of user activity is essential to detect or prevent
fraud. In this paper, we analyze data across various channels. Also, we
create a predictive model that can be used to predict such activity and
discuss how effective the model would have been to detect fraudulent
activity in the past.
Clarke Error Grid Analysis has been widely used in the accuracy
quantification of blood glucose values obtained from continuous glucose
monitoring (CGM) sensor against reference values from meter or YSI
instruments. A vivid graphic presentation of clinical accuracy of CGM sensor
data is preferred by statisticians and reviewers of regulatory agencies. SAS/
GRAPH® Annotate facility is a powerful tool for customizing, enhancing, or
changing the features of graphic outputs. Clarke Error Grid breaks down a
scatterplot of estimated glucose values versus reference values into five
zones: A, B, C, D, and E. This presentation demonstrates how to use SAS/
GRAPH, the SAS/GRAPH Annotate facility, and SAS macro applications
together to create such Error Grid for clinical accuracy determination of
CGM data against meter or YSI glucose values.
9:00 a.m.
Arrays - Data Step Efficiency
Harry Droogendyk, Stratia Consulting Inc.
Paper 519-2013
Arrays are a facility common to many programming languages, useful for
programming efficiency. SAS® data step arrays have a number of unique
characteristics that make them especially useful in enhancing your coding
productivity. This presentation will provide a useful tutorial on the rationale
for arrays and their definition and use.
9:30 a.m.
Creating ZIP Files with ODS
Jack Hamilton, Kaiser Foundation Hospitals
Paper 131-2013
ZIP files are a convenient way to bundle related files together, and can save
storage space at the same time. The ZIP format is used internally by SAS®
for SAS® Enterprise Guide® projects, but until SAS® 9.2 there was no native
way to create a ZIP file with your own SAS program. Starting in SAS 9.2, you
can create your own ZIP files using ODS PACKAGE statements. This
presentation describes how to create simple ZIP archives, and discusses
how to create an archive file with an internal directory structure.
10:00 a.m.
Three Easy Ways to Create Customized SAS® Graphs
Qinghua (Kathy) Chen, Gilead sciences Inc,
We often hear people saying that "a picture is worth a thousand words".
With that in mind, it basically tells us how powerful graphics can be when
used properly. Ways to make graphs with great visual impact has drawn a
great deal of attention from people in many fields and impactful graphics
help reviewers interpret the data. SAS® has made significant improvement
in graphs software over the past few years. With new features rolled out
such as Output Delivery System (ODS) graphics, Graphic Template
Language (GTL) and annotated data sets, creating customized graphics is as
easy as creating a simple plot. This paper will describe three easy ways to
create customized graphs in SAS.
72
11:30 a.m.
Tips for Generating Percentages Using the SAS®
TABULATE Procedure
Kathryn McLawhorn, SAS
Paper 134-2013
PROC TABULATE is one of the few Base SAS® procedures that calculate
percentages. The procedure is unique in that it has many default statistics
for generating percentages and it provides the ability to customize
denominator definitions. Determining the right denominator definition is
an important, but often challenging, aspect of calculating percentages.
Written for intermediate users, this paper discusses techniques for
enhancing PROC TABULATE output with percentage statistics. Using
examples, the paper illustrates how to add standard percentages to the
PROC TABULATE output, and it shares tips for calculating percentages that
you might have thought not possible. The paper further illustrates how to
avoid common pitfalls that are related to structuring denominator
definitions and how to format table output.
12:30 p.m.
Developer Reveals: Extended Data Set Attributes
Diane Olson, SAS
Paper 135-2013
Have you ever wanted to save non-data information with your data set?
Now you can. Extended attributes allow you to store information related to
a data set or to a particular variable in a data set. Do you want to store the
SAS® code that created a particular data set? Do you need to save a URL
that specifies information about a particular variable in your data set? Do
you want to store a description of a variable or the formula used to produce
the variable value? This presentation by the developer shows you how to
do all of that and more with extended attributes.
9:00 a.m.
10:00 a.m.
SAS® Workshop: SAS® Data Integration Studio Basics
Eric Rossland, SAS
Paper 534-2013
Paper 535-2013
This workshop provides hands-on experience using SAS Data Integration
Studio to construct tables for a data warehouse. Workshop participants will:
• define and access source data
• define and load target data
• work with basic data cleansing
11:00 a.m.
SAS® Workshop: SAS® Data Integration Studio Advanced
9:00 a.m.
Paper 536-2013
Create Your First SAS® Stored Process
Angela Hall, SAS
Paper 148-2013
Learn how to convert a simple SAS® macro into three different stored
processes! Using examples from the newly released book “50 Keys to
Learning SAS Stored Processes,” you’ll see how to build a stored process
that allows users to filter their results for the report of their dreams. You’ll
learn how to use the SAS Prompt Framework to customize your stored
process quickly and efficiently. No experience required! Suitable for
beginners. SAS® 9.2 and later.
This workshop provides hands-on experience using a combination of
DataFlux Data Management Studio and SAS® Data Integration Studio.
• Review two DataFlux Data Management Studio data jobs
• Upload the DataFlux Data Management Studio data jobs to the DataFlux
Data Management Server
• Review / create a SAS Data Integration Studio job that will execute the
uploaded data jobs on the DataFlux Data Management Server
11:00 a.m.
9:00 a.m.
So You're Still Not Using PROC REPORT. Why Not?
Ray Pass, PharmaNet/i3
Daphne Ewing, Auxilium Pharmaceuticals, Inc.
Everyone who can spell SAS® knows how to use PROC PRINT, and it
certainly has its place as a simple listing generator and as a debugging aid.
However, if a report generation/delivery tool with powerful formatting,
summarizing, and analysis features is called for, then PROC REPORT is the
solution. PROC REPORT can provide the standard PROC PRINT functionality,
but in addition, it can easily perform many of the tasks that you would
otherwise have to use the SORT, MEANS, FREQ, and TABULATE procedures
to accomplish. PROC REPORT is part of the Base SAS® product and can run
in either an interactive screen-painting mode or a batch mode. This handson workshop presents the basics of the batch (non-interactive) version of
PROC REPORT.
Ready to Become Really Productive Using PROC SQL?
Using PROC SQL, can you identify at least four ways to select and create
variables, create macro variables, create or modify table structure, and
change table content? Learn how to apply multiple PROC SQL
programming options through task-based examples. This hands-on
workshop reviews topics in table access, retrieval, structure, and content, as
well as creating macro variables. References are provided for key PROC SQL
books, relevant webinars and podcasts, and key SAS technical papers.
11:00 a.m.
FREQ Out: Exploring Your Data the Old-School Way
Stephanie Thompson, Datamum
The tried-and-true FREQ procedure just doesn’t get the attention it
deserves. But, as they say, it is an oldie but a goodie. Sometimes you just
need a quick look at your data and a few simple statistics. PROC FREQ is a
great way to get an overview of your data with a limited amount of code.
This hands-on workshop explores everything from the basic framework of
the procedure to how to customize the output. It also presents an overview
of some of the options that are available.
73
10:30 a.m.
12:00 p.m.
The Hospital Game: Optimizing Surgery Schedules to
Save Resources, and to Save Lives
SAS® Workshop: DataFlux® Data Management Studio
Basics
Paper 537-2013
This workshop provides hands-on experience using DataFlux® Data
Management Studio to profile then cleanse data. Workshop participants
will:
• learn to navigate DataFlux® Data Management Studio
• define and run a data profile
• define and run a data job
Operations Research — Room 2004
Andrew Pease, SAS
Ayesgul Peker, SAS
Paper 154-2013
Surgeons are required to perform vital operations on a daily basis, but often
their planning is not optimized for the “downstream” care. This leads to
under- or overutilization of postoperative nursing wards, which can either
compromise a hospital’s ability to provide the best possible care and ensure
the best possible patient outcome, or result in precious hospital funding
going toward underutilized resources. This paper briefly reviews some of
the academic research that is available for a data-driven, operations
research approach to solving this challenge, which is dubbed the “Hospital
Game” in some of this literature. The paper then proposes an optimizationbased approach that uses the OPTMODEL procedure to derive the best
surgery schedule for a major European hospital.
9:00 a.m.
11:00 a.m.
Using SAS® to Measure Airport Connectivity: An Analysis
of Airport Centrality in the US Network with SAS/IML®
Studio
Projecting Prison Populations with SAS® Simulation
Studio
Hector Rodriguez-Deniz, University of Las Palmas de Gran
Canaria
Pere Suau-Sanchez, Cranfield University
Augusto Voltes-Dorta, Universitat de Barcelona
The U.S. Federal Aviation Administration (FAA) estimates that $52.2 billion
will be available over the years 2011–2015 to fund airport infrastructure
developments. Because one of the main objectives is to reduce congestion
and delays, there is a need to acknowledge the importance of connectivity
(measured with a centrality indicator) when establishing funding priorities.
Currently, the FAA does not do this. In this paper, we exploit the capabilities
of SAS/IML® Studio to implement a range of centrality measures, construct
a graphical representation of the U.S. air transport network from airline
ticketing data, test the algorithms to identify hub airports, and study the
evolution of these indicators during the last decades in order to analyze the
impact of airline decisions on airport connectivity.
10:00 a.m.
Jeff Day, SAS
Ginny Hevener, NC Sentencing & Policy Advisory
Commission
Bahadir Aral, SAS
Tamara Flinchum, NC Sentencing & Policy Advisory
Commission
Emily Lada, SAS
Paper 155-2013
The majority of U.S. states are mandated to project prison populations for
the purpose of planning adequate capacity. Typical time series methods are
ineffective because they do not take into account factors like sentence
length, prior record, revocations, and legislative changes. Discrete event
simulation has proven to be a viable alternative. This paper discusses a
project in which SAS worked with the North Carolina Sentencing and Policy
Advisory Commission to build a model in SAS® Simulation Studio that
projects the number of prison beds needed for the next ten years. The
model uses current prison population data, recent court convictions,
revocations of community supervision, and estimates of growth to play out
the admissions and releases of inmates over the time horizon of the model.
Advanced Project Management beyond Microsoft
Project, Using PROC CPM, PROC GANTT, and Advanced
Graphics
Smarter Grid Operations with SAS/OR®
Paper 153-2013
Paper 156-2013
The Challenge: Instead of managing a single project, we had to craft a
solution that would manage hundreds of higher- and lower-priority
projects, taking place in different locations and different parts of a large
organization, all competing for common pools of resources. Our Solution:
Develop a Project Optimizer tool using the CPM procedure to schedule the
projects, and using the GANTT procedure to display the resulting schedule.
The Project Optimizer harnesses the power of the delay analysis feature of
PROC CPM and its coordination with PROC GANTT to resolve resource
conflicts, improve throughput, clearly illustrate results and improvements,
and more efficiently take advantage of available people and equipment.
Between the time electricity leaves utility generators and reaches your
home or business, 7% of the energy has been dissipated as heat. For the
average utility, this represents a total loss of more than $75 million each
year. Some of the electric current producing these losses does not result in
the actual production of power, and can be minimized with the proper
switching of devices that are located at strategic points in the distribution
system. These devices can also conserve energy by reducing voltages in the
distribution system and still provide a continuous supply of electricity to
the customer. This paper discusses the use of SAS/OR® to schedule device
switching to optimize the operations of the electrical distribution system.
Lindsey Puryear, SAS
Stephen Sloan, Accenture
74
11:30 a.m.
Arnie de Castro, SAS
Greg Link, SAS
12:00 p.m.
9:30 a.m.
Vehicle Retail Forecasting Demand and Inventory
Management Case Study at Shanghai General Motors
Using the ADaM ADAE Structure for Non-AE Data
Paper 157-2013
This paper describes a case study about vehicle retail forecasting demand
and inventory management in the auto industry. It describes the project's
background and the problems that were addressed using SAS®.
The final and official ADaM ADAE structure titled “Analysis Data Model
(ADaM) Data Structure for Adverse Event Analysis” was developed as an
appendix to the ADaM v2.1 to allow simple production of standard Adverse
Event tables. An ADaM sub-team is expanding this structure to cover other
data analyzed in a similar fashion, such as Concomitant Medications. The
basic premise is that data with the same analysis needs as the standard
adverse events tables can and should use this structure. This presentation,
by members of that ADaM sub-team, describes the AE analysis need and
shows to apply it for other data, such as Concomitant Medications, Medical
History, and even Laboratory Events. Examples of ADaM SAS data sets, and
useful SAS® program code are included.
Christina Zhong, shanghai general motors
12:30 p.m.
Parallel Multistart Nonlinear Optimization with PROC
OPTMODEL
Ed Hughes, SAS
Tao Huang, SAS
Yan Xu, SAS
Paper 158-2013
Nonlinear optimization has many compelling applications, including
finance, manufacturing, pricing, telecommunications, health care,
engineering, and statistics. Often a nonlinear optimization problem has
many locally optimal solutions, making it much more difficult to identify a
globally optimal solution. That’s why the multistart feature in PROC
OPTMODEL selects a number of initial points and starts optimization from
each one, significantly improving your chances of finding a global
optimum. In SAS/OR® 12.1, the multistart feature adds parallel execution.
This paper explores the multistart feature and its parallel optimization
feature, illustrating with examples drawn from research and industry.
Sandra Minjoe, Octagon Research Solutions
Mario Widel, Roche Molecular Systems
10:30 a.m.
Developing Your SDTM Programming Toolkit
David Scocca, Rho, Inc.
Paper 178-2013
Data standards such as the Study Data Tabulation Model (SDTM) make
programmer’s lives simpler but more repetitive. The similarity across
studies of SDTM domain structures and relationships presents
opportunities for code standardization and re-use. This paper discusses the
development and use of tools to simplify the process of creating SDTM data
sets, with examples of common tasks and the code to implement those
tasks. It also discusses the usefulness of a metadata system and presents a
general specification for an interface for accessing metadata. Examples
include mapping study visits, parsing dates, and standardizing test codes.
9:00 a.m.
11:00 a.m.
Easy Button: A Process for Generating Standardized
Safety- and Non-Safety- Related Clinical Trial Reports
Assessing Drug Safety with Bayesian Hierarchical
Modeling Using PROC MCMC and JMP®
Mominul Islam, Vertex Pharmaceuticals
Sanjiv Ramalingam, Vertex Pharmaceuticals Inc.
Jiannan Hu, Vertex Pharmaceuticals, Inc.
Yanwei Han, Vertex
Paper 176-2013
SAS has developed SAS® macros and template SAS programs based on its
standard (tables, figures, and listings) TFL shells for safety and non-safety
analysis. The new process includes developing reporting macros using
existing department macros to generate standard TFLs. The macros were
developed assuming the CDISC ADaM analysis data set standards, which
enable you to minimize the number of macro parameters for efficient use of
the macros by the user. The process shortens the development cycle time
and facilitates the adoption from SAS programmers to clinical reporting.
There is also a user manual and standard template programs. The process
reduces report generation time significantly and achieves the quality by
design principle.
Richard Zink, SAS
Paper 179-2013
Bayesian hierarchical models are advantageous for the analysis of adverse
events in clinical trials. First, the models can borrow strength across related
events within the MedDRA hierarchy. Second, the models can naturally
temper findings likely due to chance. We describe the implementation of
two Bayesian hierarchical models (Berry & Berry, 2004; Xia et al., 2010) used
for the analysis of adverse events using PROC MCMC. Once models are fit, it
is necessary to review convergence diagnostics to ensure that the posterior
samples of parameters sufficiently approximate the target distribution.
Numerous diagnostics are available within PROC MCMC, and we also
present a freely available JMP® add-in for MCMC (Markov Chain Monte
Carlo) dynamically interactive diagnostics, summary statistics and graphics.
12:00 p.m.
Predicting Health Care Expenditures with the MCMC
Procedure
Greg Watson, UCLA Center for Health Policy Research
Paper 496-2013
Substantial variation, excess zeros, skew and extreme outliers make fitting
and predicting health care expenditures rather difficult. This paper presents
a Bayesian model that uses the first year of the fourteenth panel
75
(2009-2010) of the nationally representative Medical Expenditures Panel
Survey (MEPS) to predict health care expenditures for individuals in the
second year. The merits of a Bayesian approach are examined and
compared to classical alternatives. Implementation in the MCMC procedure
is presented in detail, and model diagnostics and validation are discussed.
12:30 p.m.
Doctoring Your Clinical Trial with Adaptive
Randomization: SAS® Macros to Perform Adaptive
Randomization
Jenna Colavincenzo, University of Pittsburgh
Paper 181-2013
Adaptive randomization schemes have become increasingly common in
beginning stages of clinical trials and in small clinical trials. This paper
introduces two kinds of adaptive randomization schemes (treatment
adaptive randomization and covariate adaptive randomization) and
discusses the benefits and limitations of each. In addition, this paper
demonstrates how to use SAS® macros to perform these adaptive
randomization schemes in a clinical setting, and how these macros can be
modified to fit your randomization needs.
9:00 a.m.
A Macro to Verify a Macro Exists
Rick Langston, SAS
Paper 339-2013
Although the %SYSMACEXIST function can do macro existence checking, it
is limited to pre-compiled macros. This paper describes the
%MACRO_EXISTS macro, which verifies that a specified macro will be found
if invoked. The macro searches for pre-compiled macros as well as all
autocall libraries to verify their existence.
9:15 a.m.
A Simple Approach to Generate Page Numbers in X of Y
Format in ODS RTF Output
Paper 308-2013
Page numbers in X of Y format, such as "Page 18 of 280" is a common
feature of ODS RTF outputs. SAS borrows Microsoft Word processors to
compute those numbers and put them in the final output by using TITLE or
FOOTNOTE statements with "{page {\field{\fldinst{page}}} of
{\field{\fldinst{numpages}}}}" or "Page ~{thispage} of ~{lastpage}". However,
the page numbers generated by Microsoft Word processors contain field
code information displayed as "Page {PAGE \*MERGEFORMAT} of
{NUMPAGES \*MERGEFORMAT}" rather than the page numbers when Alt F9 keys are pressed. Some users such as medical writers do not like such
field code information. This paper discusses a simple way to generate page
numbers in X of Y format in ODS RTF output with the PROC REPORT
procedure.
76
9:30 a.m.
A Macro to Read in Medi-Span Text Format Database by
Data Dictionary
Sijian Zhang, VA Pittsburgh Healthcar System
Paper 344-2013
Investigators often use commercial databases to obtain useful additional
information for their researches. However, many companies do not offer
the code for transferring the data files from their deliverable file format into
the one used in the customer’s system. With many data files and variables,
the data transfer process can be very tedious. If the databases vary in
different versions, the transfer code revision can be another pain. This
paper presents an approach to simplify the data transfer process of reading
in Medi-Span drug information text data files by taking the advantage of
macro programming and its data dictionary information. One of Medi-Span
text data files, “MF2STR”, is used as an example throughout this paper.
9:45 a.m.
%GetReviews: A SAS® Macro to Retrieve User Reviews in
JSON Format from Review Websites and Create SAS®
Data Sets
Siddhartha Reddy Mandati, Oklahoma State University
Ganesh Badisa, Oklahoma State University
Paper 342-2013
The proliferation of social networking sites and consumers’ desires to create
and share content on such sites has continued to generate a huge amount
of unstructured data. Analytics users often want to tap into such
unstructured data and extract information. Many websites such as Twitter,
Facebook, and Rotten Tomatoes offer APIs for external systems to interact
and retrieve the data in JSON format. The API of Rotten Tomatoes returns
data in a complex text pattern that has information about user reviews.
Currently, there is no designated code in SAS® to read the JSON response
directly and fetch the needed data. This paper illustrates the development
and application of a SAS Macro %GetReviews to retrieve the reviews of any
desired movie from Rotten Tomatoes’ API.
10:00 a.m.
Writing Macro Do Loops with Dates from Then to Now
Ronald Fehd, retired
Paper 343-2013
Dates are handled as numbers with formats in SAS® software. The SAS
macro language is a text-handling language. Macro %do statements
require integers for their start and stop values. This article examines the
issues of converting dates into integers for use in macro %do loops. Three
macros are provided: a template to modify for reports, a generic calling
macro function which contains a macro %do loop and a function which
returns a list of dates. Example programs are provided which illustrate unit
testing and calculations to produce reports for simple and complex date
intervals.
10:15 a.m.
11:00 a.m.
On a First-Name Basis with SAS: Creating Personalized
Error Messages Using SAS 9.2
SASY Codes for Lazy people
Andrew Clapson, Statistics Canada
Valerie Hastings, Statistics Canada
Paper 352-2013
In the interest of creating a user-friendly SAS® system, you might have the
good idea to include code that checks for common errors, notifies the user,
and suggests possible solutions. Apart from simply delivering this
information to the user, you might also use customized message windows
that express congratulations upon a successful run or even deliver lighthearted finger- wagging in the case of unexpected errors. Using SAS 9.2,
this paper details the steps necessary to include basic error messaging
functionality in SAS programs. It covers notification of specific errors as well
as confirmation of successful program execution. In addition, through the
use of system macro variables, these feedback messages can surprise users
by ‘knowing’ their names and addressing them directly.
10:30 a.m.
Graph Your SAS® Off
Karena Kong, InterMune
Paper 309-2013
This paper demonstrates three different SAS® procedures for creating
graphs. For illustration purposes, the bubble plot in Figure 1, shows the
ratio of broadband users (DSL, Cable, Other) ranked by population ("List of
countries,").The data values of “Total Subscribers in Millions” and “Percent
Population Online” are annotated on the graph. The three procedures are
from SAS/GRAPH® - GPLOT, Statistical Graphics (SG) â SGPLOT and Graphics
Template Language (GTL) - PROC TEMPLATE with SGRENDER. This paper
will discuss the advantages and disadvantages between each one. Based on
the comparisons, it recommends which procedure should be used to create
a similar graph.
10:45 a.m.
Using SAS® to Assess Individuals’ Best Performance in
Multiple Dimensions
Aude Pujula, Louisiana State University
David Maradiaga, Louisiana State University
Paper 346-2013
There are many cases where we need to look at the best performance of an
individual in several disciplines over multiple time events. For instance, we
might want to know a triathlete’s best position in the three disciplines over
all the races of the season, or the highest test scores of a student in several
sub-scores. Looking at the latter example, this paper compares four
different methods implementable in Base SAS® to create a data set that
contains one record per student corresponding to the highest test scores.
Of particular interest is the use of PROC SQL combined with the SELECT
DISTINCT clause and the MAX function that allows the creation of the
desired data set in one step.
Prashanthi Selvakumar, UNT Health Science Center
Paper 353-2013
"I choose a lazy person to do a hard job, because a lazy person will find an
easy way to do it." - Bill Gates. Everyone wants to save time. While hard
work is useful, smart work is a pre requisite. Are you tired of typing codes,
then read this paper, it gives you the ways to shorten your codes. The topics
discussed in this paper include, array, do loops, macros, functions. It also
discusses the procedures and data steps where macros can save your time.
The other techniques like, combining the macros while creating html, pdf,
rtf output, to produce professional report. The possible ways of saving time
in programming are addressed in this paper.
11:15 a.m.
The SAS® Versus R Debate in Industry and Academia
Chelsea Lofland, University of California, Santa Cruz
San Luis Obispo
Paper 348-2013
Despite industry being heavily dominated by SAS®, R is used widely in
academia due to being free and open-source software that is structured
around users being able to write and share their own functions. However,
this disconnect leaves many students who are pursuing analytic degrees
struggling to get a job with less SAS experience than desired by companies.
Alternatively, they could face the struggle of transitioning everything they
learned in university from R to SAS. Ideally, one would know every possible
programming language and use the one that best suits the situation. This is
rather unrealistic. Our goal is to show the benefits of these two very
different software packages and how to leverage both of their strengths
together.
11:30 a.m.
Using SAS® to Dynamically Generate SAS® Code in Order
to Display Both Variable Label and Name as Column
Header in PROC REPORT and PROC PRINT
Victor Lopez, Baxter Healthcare Corporation
Heli Ghandehari, Baxter BioScience
Paper 349-2013
With implementation of data standards such as CDISC SDTM, datasets
contain sufficiently meaningful variable names and labels to allow direct
reporting from dataset to output (PDF, RTF, and many more). This
eliminates the necessity to program lengthy DEFINE statements in PROC
REPORT or to manually assign custom labels in PROC PRINT. This paper
illustrates an innovative approach using SAS® to dynamically generate SAS
code that enables us to solve a seemingly easy problem: displaying both
the variable label and name as a column header in PROC REPORT and PROC
PRINT.
77
11:45 a.m.
9:30 a.m.
Best Practices: PUT More Errors and Warnings in My Log,
Please!
Extending SAS® Reports to Your iPhone
Koketso Moeng, Statistics South Africa
Mary Rosenbloom, Edwards Lifesciences, LLC
Paper 350-2013
We all like to see a SAS® log that is free from errors and warnings, but did
you know that you can add your own errors and warnings to the log with
PUT statements? Not only that, but you can incorporate this technique into
your regular coding practice to check for unexpected data values. This
paper will explore the rationale and process of issuing user-created error
and warning messages to the SAS log, along with a number of examples to
demonstrate when this is useful. Finally, we will propose an upgrade to the
next version of SAS involving a user-specified keyword with its own color in
the log.
Paper 378-2013
You have jumped through all of the hoops of creating the perfect
dashboards for executives, marketing, human resources, finance, and the
project office teams, but they hardly ever get used because, frankly, your
users don't have enough time in the day to go through the reports. This is
even more true if they have to be tethered to the servers in the office to do
so. Luckily, a solution that suits the user with an iPad or iPhone is available.
Introducing Roambi—a mobile business intelligence (BI) platform that runs
on Apple's iOS platform and can be easily integrated into your existing SAS®
Enterprise BI platform.
10:00 a.m.
12:00 p.m.
The Dynamic Cube Viewer - OLAP Made Easy
The Surprisingly "Sym"ple Alternative to Hardcoding
Paper 379-2013
Rachel Carlson, Mayo Clinic
Ruchi Sharma, Mayo Clinic
Paper 351-2013
As a frequent SAS® user, do you often feel that you are spending too much
time looking up procedure results or hardcoding the values into programs?
Does your data often change causing a need to rerun analyses, forcing you
to repeat steps? Save time rerunning analysis programs by reducing the
amount of hardcoded variables and formats. Our paper will demonstrate
how to effectively use the CALL SYMPUTX routine and the SYMGET function
to make your code more flexible and minimize the possibility of data
calculation errors.
2002
9:00 a.m.
Horizontal Data Sorting and Insightful Reporting: A
Useful SAS® Technique
Justin Jia, CIBC
Amanda Lin, Bell Canada
Paper 376-2013
Sorting and ordering of data is a fundamental skill in SAS® data analysis.
Data sorting can be vertical sorting, across rows, or horizontal sorting,
across columns. Compared to vertical sort, horizontal sort is used less
frequently, and it requires the user to employ multiple sophisticated SAS
skills such as Transpose, Rotate, Array, Macro, etc. It is also an important and
useful technique for advanced data analysis and reporting in customer
profiling and metrics, which can significantly enhance the format and
layout of data reporting, and thus provide informative insights into data.
This paper will discuss the different approaches and methods of performing
horizontal sorting and presentation of SAS data, which can also expand our
horizon on data manipulation and SAS programming skills.
78
Raymond Ebben, OCS Consulting
The Dynamic Cube Viewer is a bespoke browser-based application that
offers an intuitive interface for business users to query OLAP cubes, without
the need to have an understanding of OLAP cubes. It has originally been
developed as a benchmarking tool for the Association of Dutch Insurers
and has been further developed by OCS Consulting to make it more
generic. The application reads only OLAP cube metadata and uses this to
build the user interface. An impression can be found in the attached
abstract.
10:30 a.m.
Statistical Graphics for Clinical Research Using ODS
Graphics Designer
Wei Cheng, Isis Pharmaceuticals, Inc.
Paper 380-2013
Statistical graphics play an important role across various stages in clinical
research. In this paper, I will show you the application interface and walk
you through creating some commonly used statistical graphs for clinical
research. The intended audience doesn’t need to know SAS/GRAPH®
syntax, but wants to create high-quality statistical graphs for clinical trials.
Examples will use scrambled data from real world in CDISC format.
11:00 a.m.
Visualize Your OLAP Cubes on a Map through a Stored
Process
Frank Poppe, PW Consulting
Sjoerd Boogaard, Kiwa Prismant
Paper 381-2013
What if you have an OLAP cube with a geo-dimension and you want a map
from that, but you don't have ArcGIS? Enter this general stored process. It
can read measures and dimensions from the cube, and it uses SAS/GRAPH®
software to combine that with boundary data, creating a color-coded map.
The map is clickable to navigate between the geographical levels. A
selection pane offers measures, and non-geographical dimensions surface
as (hierarchical) filters. Measures and dimensions are read from the
metadata; values for the filters are read from the cube, using MDX.
Boundaries are clipped to the right zoom level, and the picture gets a
background with roads from a web service. HTML, CSS, and JS is generated
to glue everything together and to deliver it to the portal.
11:30 a.m.
Retail — Room 3014
GTL to the Rescue!
8:00 a.m.
Paper 382-2013
Location Planning: A Look into Location Planning - Best
Practices for Resolving
Lelia McConnell, SAS
You just produced some graphs using the SGPLOT and SGPANEL
procedures. Now you want to modify the structure of your graphs in order
to make them more meaningful. You are looking for options that enable
you to split the axis values across multiple lines, add a table under a graph,
or create a template where you can conditionally execute statements and
dynamically assign variables. However, you cannot find any options within
the procedure that enable you to put these final touches on your graphs.
When all seems hopeless, ODS Graphics Template Language (GTL) comes to
the rescue!
12:00 p.m.
SCAD: Development of Statistical Information Systems
for the Provision of Census Data
Greg Pole, Statistics Centre Abu Dhabi
Paper 356-2013
The Statistics Centre - Abu Dhabi (SCAD) was founded in 2008 and seeks to
join the world’s leading statistical organizations in statistical collection,
production, and dissemination. In October 2011, SCAD conducted its first
census of population and households. In addition to using innovative
enumeration technologies (e.g., iPads), SCAD is also advancing the
development of inventive and flexible tools for accessing rich census data.
This is a positive shift towards greater access to public data in the Emirate.
The tools SCAD has developed for the 2011 Census use SAS® as a
foundation and include: on-line Thematic Mapping, on-line Community
Tables, and on-line Table Builder. These tools will be released to the Abu
Dhabi government and public in 2012, as web-based applications.
3016
12:00 p.m.
Ann Ferguson, SAS Institute
Location planning is a struggle to balance the trends of the merchandise
and stores while meeting the financial objectives of the company. The Store
Planner, plans the sales forecast and sales growth for the each and every
category available in the store or location. The number of stores in most
retail organizations is typically large and developing the store level plans is
a voluminous task. This panel discussion features recent retail trends and
efforts to maximize profits and drive improvements. Hear how these
retailers are using SAS and innovative methods and tools to develop plans
tailored for merchandise and location trends.
9:00 a.m.
Implementing Assortment Planning and the challenge of
User Adoption
Ann Ferguson, SAS Institute
Merchandise Assortment planning plays a pivotal role in creating and
maintaining profitability. No other area within a retail business has such a
direct impact on bottom line profit (or loss). It is, therefore, crucial that
merchandisers have a broad understanding of the best practice approaches
that have evolved and continue to evolve in order that they are able to
optimize the financial return on the investment that is under their control.
10:00 a.m.
Reclass: What Does It Mean to You!
Amy Clouse, Dick's Sporting Goods
“Google-Like” Maps in SAS®
Reclass: What does it mean to you! Reclass can be a daunting task to take
on in your organization. This session will focus on best practices for
preparation and execution, compiled from several SAS® customers.
Paper 377-2013
11:00 a.m.
Darrell Massengill, SAS
We are frequently asked if we can have maps similar to Google Maps in
SAS®. Customers want the background image displayed behind their data
so they can see where streets or other features are located. They may also
want to pan and zoom the map. Unfortunately, Google has legal
restrictions and limitations on the use of their maps. Now, you can have
“Google-like” maps inside of SAS. You may have already seen this capability
in products like SAS® Visual Analytics Explorer and other products using
them will be available in future releases. This presentation will discuss and
demonstrate these new capabilities in SAS Visual Analytics Explorer, SAS/
GRAPH®, and other products.
Forecasting to Support Planning
Julie Rankin, Belk
Strength in Numbers: Using Demand Forecasting to Drive Merchandise and
Store Performance Hear how these retailers are using SAS Demand
Forecasting for Retail to strengthen their numbers through advanced retail
management from the SAS forecast engine in their planning processes.
With broad assortment and locations, diverse consumer behavior and price
fluctuations, predicting demand and performance can be quite a challenge
in the pre-season and in-season planning processes. These retailers are
using analytics to drive results and supplement the art of the merchant
expertise to generate efficiencies via a scientific approach to forecasting
demand. Forecast results drive better business decisions, improved
planning processes and forecast accuracy. The user experiences reflect
streamlined processes, improved productivity and better demand patterns.
79
12:00 p.m.
SAS® Retail Planning 7.2 Demonstration
Elaine Markey, SAS
Paper 396-2013
A live demonstration of SAS Retail Planning 7.2. This demonstration will
show how this release helps retailers plan their financials and organize
customer-centric localized assortments in an effective and efficient manner.
SAS® Enterprise Guide® Implementation and
Usage — Room 3002
9:00 a.m.
SAS® Enterprise Guide®, Best of Both Worlds: Is it Right
for You?
SAS and Big Data — Room 3001
9:00 a.m.
High Performance Statistical Modeling
Bob Rodriguez, SAS
Robert Cohen, SAS
Paper 401-2013
The explosive growth of data, coupled with the emergence of powerful
distributed computing platforms, is driving the need for high-performance
statistical modeling software. SAS has developed a series of procedures that
perform statistical modeling and model selection by exploiting all of the
cores available – whether in a single machine or in a distributed computing
environment. This presentation includes a demonstration of the current
capabilities of this software as well as guidance on how and when these
high-performance procedures will provide performance benefits.
10:00 a.m.
SAS® Visual Analytics Road Map
Greg Hodges, SAS
Paper 513-2013
Journey down the SAS® Visual Analytics road map with product
management! Learn about the reporting, exploration and even new data
visualizations under consideration for future releases.
12:00 p.m.
Getting Started with SAS® Visual Analytics
Administration and Deployment
Meera Venkataramani, SAS
Gary Mehler, SAS
Paper 515-2013
Discover invaluable tips on how to get started with SAS® Visual Analytics
deployment, administration and monitoring. Hear tips and best practices
for a flawless implementation.
80
Whether you are new to SAS® or a seasoned SAS Programmer, you still face
the same dilemma. Does SAS® Enterprise Guide® represent the best of both
worlds to make the transition to SAS easier with a point-n-click interface or
enhance your productivity with over 90 tasks? Do you follow the same
traditional path taken by millions who learned SAS many decades ago or do
you take the yellow brick road to directly analyze your data? This
presentation explores the vast differences between these two cultures and
how they impact your programming environment. While there are
numerous benefits to using SAS Enterprise Guide, there are also some
caveats to keep in mind to make the transition smoother.
10:00 a.m.
Update, Insert, and Carry-Forward Operations in
Database Tables Using SAS® Enterprise Guide®
Paper 417-2013
You want to use SAS® Enterprise Guide® to simulate database logic that
includes any of: update, insert, carry-forward operations on old, changed, or
new rows between two data sets, to create a new master data set. However,
the Query Builder Task GUI does not have an Update/Insert option.
Methods for simple types of update, insert, and/or carry-forward operations
are described and illustrated using small data sets. First, we review Base
SAS® methods, including DATA step and PROC SQL code. Then, two GUIonly/Task-based methods are described: one based on the Sort Data Task
GUI; the other on the Query Builder Task GUI. The issue of whether integrity
constraints are preserved is also discussed.
10:00 a.m.
Update, Insert, and Carry-Forward Operations in
Database Tables Using SAS® Enterprise Guide®
Sreenivas Mullagiri, iGATE Global Solution
Paper 417-2013
You want to use SAS® Enterprise Guide® to simulate database logic that
includes any of: update, insert, carry-forward operations on old, changed, or
new rows between two data sets, to create a new master data set. However,
the Query Builder Task GUI does not have an Update/Insert option.
Methods for simple types of update, insert, and/or carry-forward operations
are described and illustrated using small data sets. First, we review Base
SAS® methods, including DATA step and PROC SQL code. Then, two GUIonly/Task-based methods are described: one based on the Sort Data Task
GUI; the other on the Query Builder Task GUI. The issue of whether integrity
constraints are preserved is also discussed.
10:30 a.m.
A Comparison between GUI Prompts of SAS® Enterprise
Guide® 4.1 and 4.3 and Approaches for Developing NextGeneration Prompts
9:00 a.m.
Menaga Ponnupandy, Technosoft Corp
Paper 418-2013
SAS® codes have to be edited when the criteria of execution changes. The
use of GUI Prompts helps in preserving source code from changes and in
automation of SAS® Enterprise Guide® Projects by passing run-time
parameters to SAS. The main purpose of this paper is to compare the
advanced features or functionalities of GUI Prompts between SAS
Enterprise Guide 4.1 and SAS Enterprise Guide 4.3. This paper also discusses
about the limited ability of prompts and provides tips for handling such
situations or highlights the need for development of future generation
prompts.
11:00 a.m.
Making do with less: Emulating Dev/Test/Prod and
Creating User Playpens in SAS® Data Integration Studio
and SAS® Enterprise Guide®
David Kratz, d-Wise
Paper 419-2013
Have you ever required a Dev / Test / Prod environment but found yourself,
for whatever reason, unable to lay down another SAS Installation? Have you
ever discovered that your results have been overwritten by a team
member? Our ability to use SAS is shaped by the environment in which the
software is installed, but we often don't have as much control over that
environment as we'd like. However, we can often emulate the setup we'd
prefer by configuring the one we have. This paper explores this concept
using techniques which can be applied to development in SAS Data
Integration Studio and SAS Enterprise Guide.
11:30 a.m.
What SAS® Administrators Should Know About Security
and SAS® Enterprise Guide®
Casey Smith, SAS
Paper 420-2013
SAS® Enterprise Guide® is a flexible and powerful tool in the hands of your
users. However, as every superhero knows, with power comes
responsibility. As an administrator, you want to ensure various groups of
users have access to the specific resources they need to be as productive as
possible, while at the same time protecting company assets and minimizing
risk. This paper explores various security considerations from the SAS
Enterprise Guide perspective, such as authentication, authorization, user
administration, access management, encryption and role-based availability
of application features.
Estimating Censored Price Elasticities Using SAS/ETS®:
Frequentist and Bayesian Approaches
Christian Macaro, SAS
Jan Chvosta, SAS
Kenneth Sanford, SAS
James Lemieux, SAS Institute
Paper 448-2013
The number of rooms rented by a hotel, spending by “loyalty card”
customers, automobile purchases by households—these are just a few
examples of variables that can best be described as “limited” variables.
When limited (censored or truncated) variables are chosen as dependent
variables, certain necessary assumptions of linear regression are violated.
This paper discusses the use of SAS/ETS® tools to analyze data in which the
dependent variable is limited. It presents several examples that use the
classical approach and the Bayesian approach that was recently added to
the QLIM procedure, emphasizing the advantages and disadvantages that
each approach provides.
9:00 a.m.
Considerations and Techniques for Analyzing Domains
of Complex Survey Data
Taylor Lewis, U.S. Office of Personnel Management
Despite sounding like a straightforward task, making inferences on a
domain, or subset, of a complex survey data set is something that is often
done incorrectly. After briefly discussing the features constituting complex
survey data, this paper explains the risks behind simply filtering the full data
set for cases in the domain of interest prior to running a SAS/STAT® survey
procedure such as PROC SURVEYMEANS or PROC SURVEYREG. Instead, it
shows how one should use the DOMAIN statement or create a domainspecific analysis weight. Also discussed in detail are considerations and
approaches to the very common objective of testing whether the
difference between two domain means is statistically significant.
10:00 a.m.
Markov Chains and Zeros in My Data: Bayesian
Approaches in SAS® That Address Zero Inflation
Matthew Russell, University of Minnesota
Paper 450-2013
In recent releases of SAS/STAT® software, a number of procedures that
perform Bayesian methodologies have been incorporated. A common
modeling problem across many disciplines is that of addressing largerthan-expected proportions of zeros, a problem that is exacerbated when
counts and probabilities of zeros are heterogeneous. This paper uses
examples from the ecological literature to perform Bayesian analyses on
discrete data with zero inflation. We focus primarily on the MCMC
procedure, but also address use of Bayesian methods in the FMM and
81
GENMOD procedures. We fit zero-inflated models under conditional
binomial, Poisson, and negative binomial assumptions both with and
without random intercept effects.
illustrates the basic framework of an LGC model and introduces a SAS
macro, %LGCM, that fits a latent growth model and computes incremental
fit indices based on more appropriate baseline models.
10:00 a.m.
11:00 a.m.
Exploring Health Trends and Risk Behavior Analysis in
American Youth Using PROC SURVEYFREQ and PROC
SURVEYLOGISTIC
The Box-Jenkins Methodology for Time Series Models
Deanna Schreiber-Gregory, North Dakota State University
Paper 451-2013
This study looks at recent health trends and behavior analyses of youth in
America. Data used in this analysis was provided by the Centers for Disease
Control and Prevention and gathered using the Youth Risk Behavior
Surveillance System (YRBSS). This study outlines demographic differences
in risk behaviors, health issues, and reported mental states. Interactions
between risk behaviors and reported mental states were also analyzed.
Visual representations of frequency data for the national results are also
provided and discussed. A final regression model including the most
significant contributing factors to suicidal ideation is provided and
discussed. Results included reporting differences between the years 1991
and 2011. All results are discussed in relation to current youth health trend
issues. Data was analyzed using SAS® 9.3.
Theresa Ngo, Warner Bros. Home Entertainment
Paper 454-2013
A time series is a set of values of a particular variable that occur over a
period of time in a certain pattern. The most common patterns are
increasing or decreasing trend, cycle, seasonality, and irregular fluctuations
(Bowerman, O’Connell, and Koehler 2005). To model a time series event as a
function of its past values, analysts identify the pattern with the assumption
that the pattern will persist in the future. Applying the Box-Jenkins
methodology, this paper emphasizes how to identify an appropriate time
series model by matching behaviors of the sample autocorrelation function
(ACF) and partial autocorrelation function (PACF) to the theoretical
autocorrelation functions. In addition to model identification, the paper
examines the significance of the parameter estimates, checks the
diagnostics, and validates the forecasts.
11:00 a.m.
10:30 a.m.
Introducing the New ADAPTIVEREG Procedure for
Adaptive Regression
Forecasting Net Job Creation Using SAS®
Casey Sperrazza, University of Alabama
Paper 453-2013
Using data from the U.S. Census Bureau’s Business Dynamics Statistics, net
job creation is forecast economywide and by sector. Forecasts are carried
out economywide using exponential smoothing and ARIMA models.
Forecasting is carried out by Census Bureau–defined sectors using ARIMA
models. Data are from 1977–2010, and net job creation is forecast through
2020.
10:30 a.m.
Modeling Change over Time: A SAS® Macro for Latent
Growth Curve Modeling
Pei-Chin Lu, University of Northern Colorado
Robert Pearson, University of Northern Colorado
Paper 452-2013
In recent years, latent growth curve (LGC) modeling has become one of the
most promising statistical techniques for modeling longitudinal data. The
CALIS procedure in SAS® 9.3 could be used to fit an LGC model. As one
application of structural equation modeling (SEM), LGC modeling relies on
indices to evaluate model fit. However, it has been pointed out that when
you are obtaining incremental fit indices, the default baseline model used
in many popular SEM software packages, including PROC CALIS, is generally
not appropriate for LGC models (Widaman and Thompson 2003). This paper
82
Warren Kuhfeld, SAS
Weijie Cai, SAS
Paper 457-2013
Predicting the future is one of the most basic human desires. In previous
centuries, prediction methods included studying the stars, reading tea
leaves, and even examining the entrails of animals. Statistical methodology
brought more scientific techniques such as linear and generalized linear
models, logistic regression, and so on. In this paper, you will learn about
multivariate adaptive regression splines (Friedman 1991), a nonparametric
technique that combines regression splines and model selection methods.
It extends linear models to analyze nonlinear dependencies and produce
parsimonious models that do not overfit the data and thus have good
predictive power. This paper shows you how to use PROC ADAPTIVEREG (a
new SAS/STAT® procedure for multivariate adaptive regression spline
models) by presenting a series of examples.
11:30 a.m.
Exploring Time Series Data Properties in SAS®
David Maradiaga, Louisiana State University
Aude Pujula, Louisiana State University
Hector Zapata, Louisiana State University
Paper 456-2013
Box and Jenkins popularized graphical methods for studying time series
properties of time series data. Dickey and Fuller did the same for unit root
tests. Both methods seek to understand the nonstationary properties of
improvement in overall throughput and has allowed eBay Inc in ~30%
additional processing capacity, and thereby enabled evaluating more
experiments.
data, and SAS® software is a popular tool used by applied researchers. The
purpose of this paper is to provide a series of steps using the SAS macro
language, PROC SGPLOT, PROC ARIMA, PROC AUTOREG, and the %dftest
macro to diagnose nonstationary properties of data. A comparison of three
competing SAS procedures is presented, with SAS capabilities highlighted
using simulated time series.
10:30 a.m.
12:00 p.m.
Hardening a SAS® Installation on a Multi Tier installation
on Linux
Nontemporal ARIMA Models in Statistical Research
David Corliss, Magnify Analytic Solutions
Mathematical models employing an autoregressive integrated moving
average (ARIMA) have found very wide applications following work by Box
and Jenkins in 1970, especially in time series analysis. ARIMA models have
been very successful in financial forecasting, forming the basis of such
things as predicting how much gas prices will rise. However, no
mathematical requirement exists requiring the data to be a time series: only
the use of equally spaced intervals for the independent variable is
necessary. This can be done by binning data into standard ranges, such as
income by $10,000 intervals. This paper reviews the fundamental statistical
concepts of ARIMA models and applications of non-temporal ARIMA
models in statistical research. Examples and applications are given in
biostatistics, meteorology, and econometrics as well as astrostatistics.
Jan Bigalke, Allianz Managed Operations & Services SE
Paper 481-2013
The security requirements of today require in some use cases the hardening
of a SAS® Installation. This paper describes the practical steps of securing
the SAS web applications and the impact to the Base SAS® Services on the
SAS computer tiers. The SAS® Enterprise BI Server will be the object of this
explanation. The principles of a secure architecture will be described and
the options to secure the individual components presented.
11:00 a.m.
Do I Need a Migration Guide or an Upgrade Coach?
Donna Bennett, SAS
Mark Schneider, SAS
Gerry Nelson, SAS
Paper 482-2013
2006
9:00 a.m.
Benchmarking SAS® I/O: Verifying I/O Performance
Using fio
Spencer Hayes, J. S. Hayes, Inc.
Paper 479-2013
Input/Output (I/O) throughput is typically the most important computing
aspect of a SAS® environment. Bandwidth requirements ranging from
25MB/sec/core to 135MB/sec/core are common in a high-performance SAS
system. Insuring that the storage subsystem can meet the demands of SAS
is critical to delivering the performance required by the business and user
community. Ideally, SAS administrators could run real-world SAS jobs to
benchmark the I/O subsystem. However, technical and logistical challenges
frequently make that option impractical. The open source software tool
called “fio” provides a method for accurately simulating I/O workloads. It is
configurable to match closely the existing or expected I/O profile for a SAS
environment.
9:30 a.m.
eBay Quadruples Processing Speed with SAS® InDatabase Analytics for Teradata
John Scheibmeir, eBay
Don’t bring a hammer when you need a paintbrush! There are many kinds
of changes you can make to your SAS® deployment. Sometimes, SAS
migration tools may provide the best path for making changes to your
system. Other times, your changes may need other deployment and
management tools. Whether you are an administrator managing the
changes or an IT administrator overseeing SAS software, this paper will help
you choose the right tools to plan and manage SAS software changes. As an
added bonus, the paper includes a glossary of common terms and concepts
that often require collaboration between IT and SAS management.
11:30 a.m.
Integrating SAS® into Your Operational Environment:
SOA: A Means to an End
Saravana Chandran, SAS
Rob Stephens, SAS
Paper 483-2013
The impact of business analytic models has proven value for enterprises.
Many SAS customers have highly valuable analytical assets, ranging from
analytical models to analytical services specific to domain. We have seen a
significant stream of requests for assistance in taking the next step to
deploy these models into their primary operational business applications.
Service-oriented architecture (SOA) is an architectural style designed to
enable flexibility, reusability and interoperability; it provides one of the
primary means for integrating SAS® with your operational application
environment. The paper walks through the integration, runtime
environment, governance and best practices all in the context of SOA and
SAS Business Analytics.
Working efficiently with HUGE data sets consisting of millions of rows and
hundreds of columns summing up to gigabytes of storage is a challenge
that many users and organizations face today. In addition to processing
large amounts of data, additional constraints include end-to-end
processing time, implications of transfer of processing to the database,
storage space, system resources, data transfer, etc. Utilizing SAS® indatabase processing on eBay’s Teradata based Singularity Platform has
reduced end-to-end processing time by a factor of 4 at eBay Inc. This
83
12:00 p.m.
How to Choose the Best Shared File System For Your
Distributed SAS® Deployment
Barbara Walters, SAS
Ken Gahagan, SAS
Leigh Ihnen, SAS
Vicki Jones, SAS
Paper 484-2013
A shared file system is an integral component of all SAS® Grid Manager
deployments, SAS Enterprise BI deployments with load balanced servers on
multiple systems, and other types of distributed SAS applications. This
paper explains how SAS software interacts with the file system and how a
shared file system behaves differently than a non-shared file system. It
describes the factors that determine whether a particular file system is a
good fit for your environment and how to choose the file system that best
meets your needs.
12:30 p.m.
Bridging the Gap Between SAS® Applications Developed
by Business Units and Conventional IT Production
Euwell Bankston, Union Bank, NA
Paper 485-2013
Multiple factors are involved in the decision by an enterprise to decide
whether to allow a business unit to run its own production versus having
SAS® applications developed by business units run in conventional IT
production. There can be a wide gap between the business unit view of
"production-ready" programs vs. core IT standards for production systems.
The nature of the gap is discussed here, and also the risks of business-run
production. Specific suggestions are made regarding whether IT and
business should have joint ownership of critical SAS applications vs.
segregated roles, and when/how should SAS-based systems be migrated
into a fully controlled IT production environment.
84
A
Aanderud, Tricia 8 , 16 , 61 , 73
Abousalh-Neto, Nascif 41
Achanta, Bhargav 21
Agrawal, Gaurav 41
Akerman, Meredith 26
Albright, Russell 44 , 59
Alexander, Malcolm 8 , 42
Amezquita, Darwin 43
Anderson, Brett 16
Anderson, Marty 58
Ansari, Taufique 28
Aral, Bahadir 74
Atassi, Sam 8
Atkinson, Chad 45
Auyuen, Wuong Jodi 53
Axelrod, Elizabeth 56
Azimaee, Mahmoud 41
B
Badisa, Ganesh 76
Baecke, Philippe 63
Baer, Darius 71
Bailey, Jeff 42
Balan, Tonya 32
Balinskaite, Violeta 64
Ball, Kathy 20
Bankston, Euwell 84
Barnes, Arila 45
Battaglia, Michael 17
Battiston, Christopher 22 , 31 , 57
Beatrous, Steve 6
Beaty, Brenda 13
Beaver, Allan 58
Beaver, James 7
Beaver, Richard 12
Becker, Matthew 51
Bedford, Denise 45
Bee, Brian 10
Bellara, Aarti 19 , 62
Bell, Bethany 23 , 2
Benjamin, William 37 , 56
Bennett, Donna 10 , 83
Bentley, John 8 , 46
Berryhill, Tim 47
Betsinger, Alicia 41
Bhardwaj, Pankaj 20
Bibb, Barbara 16
Bieringer, Alicia 59
Bigalke, Jan 83
Billings, Thomas 55 , 80 , 84
Birds, Andy 34
Bjurstrom, Jennifer 62
Bogard, Matt 7
Bolen, Tison 33
Bonham, Bob 35
Boniface, Christopher 7 , 54
Bonney, Christine 14
Boogaard, Sjoerd 78
Boorse, Carrie 50
Booth, Allison 57
Boscardin, John 33
Bost, Christopher 55 , 56
Bouedo, Mickael 6
Bretz, Michael 67
Bright, Terri 17
Brooks, Sandra 52
Brown, James R 54
Brown, Mark 71
Brown, Tony 67
Busi Reddy, Srinivas Reddy 24
C
Cai, Weijie 82
Carey, Delicia 23
Carlson, Rachel 78
Carpenter, Art 5 , 12 , 69
Carr, Brenda 58
Cathie, Andrew 37
Cegielski, Paul 7
Cenzer, Irena 33
Chadha, Rajbir 19 , 53
Chakraborty, Goutam 10 , 21 , 22 ,
23 , 24 , 25 , 28 , 37 , 38 , 53 , 71 , 76
Chandran, Saravana 31 , 83
Chapman, Don 7
Chen, Fang 2
Cheng, Alice 13
Cheng, Wei 78
Chen, Min 20
Chen, Qinghua (Kathy) 72
Chen, Wei 71
Chen, Yi-Hsin 18 , 19 , 62
Cheong, Michelle 9 , 21
Chew, Maureen 40
Chick, Brian 71
Chitale, Anand 41 , 61
Choy, Justin 32 , 70
Choy, Murphy 9 , 21 , 29
Christian, Stacey 47 , 71
Chung, Kevin 53
Chvosta, Jan 81
Clapson, Andrew 77
Clouse, Amy 59 , 79
Clover, Lina 61
Cochran, Ben 38
Cohen, Robert 80
Colavincenzo, Jenna 76
Coleman, Oita 7
Conley, Lisa 33
Conway, Ted 27
Corliss, David 40 , 83
Correa Bahnsen, Alejandro 43
Cox, James 59
Craig, Jean 15
Crain, Charlotte 42
Crawford, Peter 10 , 66
Crevar, Margaret 67
Cui, Xiangchen 20 , 75
Culley, Joan 15
Cunningham, John 8
Czika, Wendy 44
D
D'Agostino, Lucy 64
Darade, Harshal 27
Darden, Paul 26
Davies, Jennifer 30
Day, Eric 27
Day, Gavin 12 , 13
Day, Jeff 74
de Castro, Arnie 74
Deaton, Randall 56
Deguire, Yves 5
DelGobbo, Vince 48
Delwiche, Lora 47
Derby, Nate 52 , 54
Derr, Bob 65
deVille, Barry 45
Dhillon, Rupinder 12
Dickey, David 62
Dickinson, L. Miriam 13
Dingstad, Bernt 40
Ding, Sheng 26
Dong, Qunming 20
Dong, Tianxi 25
Donovan, Bill 14
Dorfman, Paul 47
Dowling, Molly 23
Droogendyk, Harry 10 , 72
Drutar, Michael 7
Dunn, Vann 14
Duraidhayalu, Hari harasudhan 24 ,
25
E
Ebben, Raymond 78
Eberhardt, Peter 6 , 11 , 19
Eckler, Lisa 22 , 30
Edwards, David 51
Elkin, Eric 13 , 2
Elliott, Alan 27
Ellis, Dylan 39
Ene, Mihaela 23 , 2
Erdman, Donald 47 , 71
Eubanks, Elizabeth 24
Ewing, Daphne 73
F
Faria, Plinio 41
Fecht, Marje 10 , 12
Fehd, Ronald 29 , 76
Ferguson, Ann 79
Flam Zalcman, Rosely 21
Flavin, Justina 13
Flinchum, Tamara 74
Florez Hormiga, Deybis 44
Fogleman, Stanley 28
Ford, Bill 12
Forst, Fred 66
Frost, Mike 42
Fu, I-kong 61
Fu, Yu 23 , 25 , 28 , 53
G
Gaethe, Gary 22
Gahagan, Ken 84
Galinson, Boaz 46
Gao, Nanxiang 43
Gao, Yubo 18
Gao, Yun 63
Gardiner, Joseph 3
Garla, Satish 37 , 51
Garrett, Guy 69
George, Tammi Kay 7
Ghadban, Khaled 12
Ghandehari, Heli 28 , 77
Ghanekar, Saurabh 25
Gibbs, Phil 3
Gibbs, Sandy 10
Gidley, Scott 42 , 43
Gilsen, Bruce 39 , 55
Glassman, Amy 70
Goldman, Robert 57
Gonzalez, Andres 43
Goodin, Shelly 14
Goodman, Melody S. 64
Gordon, Leonard 10
Greenfield, Jason 33
Gregory, Raymond 14
Greni, Chris 55
Gribbin, Dave 70
Grover, Swati 22
Gu, Haoyu 29
Guido, Lori 19 , 67
Gunes, Funda 1
Guo, Baojian 26
Gupta, Saurabh 59
Gupta, Sunil 1 , 30 , 73 , 80
Gutierrez, Bobby 33
H
Hadden, Louise 17
Haigh, Doug 66
Hale, Jessica 26
Hall, Angela 8 , 16 , 73
Haller, Susan 44
Hamilton, Jack 35 , 72
Hamm, Rob 35
Hamstra, Kirsten 14
Han, Yanwei 75
Hardin, J. Michael 32
Harris, Kriss 21
85
Hastings, Valerie 77
Hatcher, Diane 35
Hatton, Robert 45
Hayes, Spencer 83
Heaton, John 36 , 42
Hebbar, Prashant 58
Hecht, Michael 37
Heiliger, Carsten 71
Heiney, Sue 65
Hembree, Tina 52
Hemedinger, Chris 5 , 61
Hendel, Russell 29
Henderson, Don 49
Henrick, Andrew 47
He, Tao 50
Hevener, Ginny 74
High, Robin 65
Hill, Aaron 60
Hill, Melissa 15
Hill, Toby 29
Hinson, Joseph 6
Hodges, Greg 7 , 80
Holland, Philip 69
Holman, James 7
Holmes, Harold 62
Holmes, Steven 40
Homes, Paul 66
Hopping, Albert 51
Horwitz, Lisa 52
Howell, Andrew 52
Hoyle, Larry 54
Huang, Chao 23
Huang, Tao 75
Huang, Zhongwen 17 , 50
Hughes, Ed 75
Hu, Jiannan 75
Hummel, Andrew 57
Hung, Dorothy 51
Hunley, Chuck 35
Hunter, Tim 69
Hurley, George 56
Hynan, Linda 27
Jiang, Hong 40
Jiang, Shaozong 43
Ji, Junyao 45
John, Christine 26
Johnson, Maribeth 1
Jones, Vicki 84
Jordan, Lori 70
Jordan, Mark 2
K
Ihnen, Leigh 84
Inbasekaran, Rajesh 25
Inoa, Ignacio A. 64
Islam, Mominul 75
Ito, Shunsuke 14
Izrael, David 17
Kachare, Atul 58
Kaduwela, Vijitha 25
Kakde, Deovrat 25
Kalgotra, Pankush 38
Kalidindi, Shiva 53
Karafa, Matt 6
Karimova, Mariya 26
Karnavat, Pranav 24
Kauhl, Justin 9
Keefer, Tom 34
Keener, Gordon 32
Keith, Jr., Michael 14
Kellermann, Anh 19
Kelley, David 57
Kelly, Patrick 56
Kenkare, Pragati 51
Kent, Paul 12 , 13 , 40 , 60
Keshan, Xia 39 , 56
Kezik, Julie 15
Khalil, Yehia 52
Khambhati, Shilpa 16
Khanna, Kamya 55
Kiernan, Kathleen 3
Kilburn, Julie 54
Kim, Eun Sook 19
Kim, Jonghyun 25
Kincaid, Chuck 48
King, Michael 35
Kirby, Katharine 33
Kong, Karena 77
Konya, Mark 20
Koval, Scott 69
Kratz, David 81
Kromrey, Jeffrey 18 , 19 , 20 , 62
Krycha, Karl 9
Kuhfeld, Warren 2 , 82
Kukhareva, Polina 1
Kuligowski, Andrew 49
Kunselman, Thomas 61
Kurtz, Joy 18
J
L
Jacob, Jeffin 22
Jansen, Lex 18
Javanainen, Jouni 66
Jia, Justin 78
Jiang, Charley 5
Jiang, Haibo 60
La Valley, Richard 18 , 45
Lada, Emily 74
Lafler, Kirk Paul 3 , 15 , 18 , 49 , 78
Lai, Ginny 13
Lan, Bob 15
Landy, David 18
I
86
Langston, Julianna 57
Langston, Rick 39 , 76
Lau, Miu Ling 19
Lawhorn, Bari 69
Lee, Taiyeong 44
Lemieux, James 81
Lepkowski, James 5
Leslie, Elizabeth 27
Leslie, Scott 60
Lesser, Martin 26
Levine, Stuart 50
Lewis, Taylor 81
Li, Arthur 42 , 47
Li, Isaac 62
Li, Jian 43
Liming, Douglas 54
Lin, Alec 44
Lin, Amanda 78
Lin, Guixian 32
Link, Greg 74
Lin, Zhangxi 25 , 27
Li, Regan 15
Li, Siming 27
Li, Suwen 15
Liu, Charlie 30
LIU, FENG 61
Liu, Jiawen 22
Lix, Lisa 27
Li, Yahua 43
Li, Zhiyong 35
Lofland, Chelsea 77
Loman, Cynthia 15
Lopez, Adolfo 21
Lopez, Victor 28 , 77
Lou, Youbei 17
Low, Ronald 14
Ludwig, David 18
Lund, Bruce 39
Lund, Pete 39
Lu, Pei-Chin 82
Lycan, Christiana 8
M
Macaro, Christian 81
Macfarlane, Andrew 35
Maier, Mike 38
Maldonado, Miguel 44
Mandati, Siddhartha Reddy 21 , 76
Manickam, Airaha Chelvakkanthan
20 , 61
Mann, Robert 21
Maradiaga, David 77 , 82
Markey, Elaine 80
Massengill, Darrell 79
Ma, Sai 15
Matange, Sanjay 11 , 13 , 56 , 58
McCann, Claudia 43
McConnell, Lelia 79
McGahan, Colleen 54
McGaugh, Miriam 25 , 53
McGowan, Kevin 40
McLawhorn, Kathryn 72
McNeill, Bill 6
Meeran Mohideen, Musthan Kader
Ibrahim 22 , 24
Mehler, Gary 80
Mendelsohn, Andy 40
Mengelbier, Magnus 51
Meng, Xiangxiang 44
Miao, Yinghui 33
Miller, Tracie 18
Minjoe, Sandra 11 , 75
Miralles, Romain 55
Misra, Aditya 45
Mistler, Stephen 64
Mistry, Jugdish 31
Moeng, Koketso 78
Monaco, Rick 51
Moore, Stephen 19 , 67
Moreno-Simon, Tawney 28
Morton, Steve 69
Mu, George 13
Mullagiri, Sreenivas 80
Muller, Roger 25 , 26
Mulugeta, Dawit 33
Murphy, William 55
Myers, Keith 70
Myers, Susan 57
N
Na, Beeya 2
Nagarajan, Srihari 24
Nash, Michael 50
Nauta, Frank 58
Nelson, Gerry 83
Nelson, Greg 51 , 66
Nelson, Robert 23
Ngo, Theresa 82
Ng, Song Lin 22
Nguyen, Diep 19
Nguyen, Mai 16 , 24
Nisbet, Stuart 7
Nist, Pauline 13
Nizam, Azhar 30 , 57
Noga, Steve 15
Nori, Murali 70
O
O'Connor, Dan 38
O'Neil, Michael 43
Okerson, Barbara 17 , 31
Olson, Diane 72
Olson, Mike 12 , 13
Ong Yeru, Cally 9
Orange, Cary 71
Osborne, Anastasiya 30
Osborne, Mary 59
Ottesen, Rebecca 54 , 56 , 77
Otto, Greg 42
Overby Wilkerson, Amy 16
Overton, Stephen 8
P
Pakalapati, Tathabbai 20
Palaniappan, Latha 50
Pallone, Mark 5
Pan, Helen 65
Pan, Minghua 43
Pantangi, Anil Kumar 21
Parker, Chevell 38
Parks, Jennifer 66 , 67
Pass, Ray 73
Pasta, David 32
Pearson, Robert 82
Pease, Andrew 74
Pedersen, Casper 9
Peker, Ayesgul 74
Pelan, Margaret 58
Peravalli Venkata Naga, Krutharth
Kumar 72
Peters, Amy 35 , 65
Peterson, Jared 45
Petrova, Tatyana 42
Pham, Hung 7 , 54
Pham, Thanh 18
Pletcher, Rich 34
Plumley, Justin 59
Polak, Leonard 29
Pole, Greg 79
Ponnupandy, Menaga 81
Poppe, Frank 78
Potter, Ken 45
Poulsen, Rachel 65
Priest, Elisa 26
Prins, Jared 45
Pujula, Aude 77 , 82
Punuru, Janardhana 44
Purushothaman, Ramya 24 , 53 , 61
Puryear, Lindsey 74
Q
Qi, Lingxiao 43
Qin, Xiaojin 6
Qu, Hengrui 16
Qureshi, Lubna 51
R
Raimi, Steven 39
Raithel, Michael 1 , 46 , 66
Rajamani, Mythili 55
Ramalingam, Sanjiv 75
Rankin, Julie 79
Rasmussen, Patrice 18 , 62
Ratcliffe, Andrew 37 , 58
Rausch, Nancy 8 , 42 , 43
Ravuri, Sahithi 21
Rayabaram, Srikar 23 , 24 , 72
Ray, Robert 32
Reay, Stefanie 14
Redpath, Christopher 41
Rey, Timothy 9
Rhodes, Dianne Louise 52
Richardson, Brad 6
Richardson, Kari 11 , 12 , 73 , 74
Riddiough, Christine 1
Rigden, Chris 34
Rittman, Sarah 51
Rodriguez de Gil, Patricia 18 , 19 ,
20 , 62
Rodriguez-Deniz, Hector 74
Rodriguez, Bob 32 , 80
Roehl, William 5
Rogers, Stuart 67
Romano, Jeanine 18 , 62
Rosenbloom, Mary 5 , 78
Rossland, Eric 2 , 11 , 12 , 47 , 48 ,
49 , 73
Royal, Annette 41
Russell, Matthew 81
Ryan, Laura 44
S
Sadhukhan, Shreya 28
Sall, John 1
Sams, Scott 40
Sanders, Scott 58
Sanford, Kenneth 81
Sarkar, Deepa 55
Sarkar, Mantosh Kumar 10 , 22
Sauer, Brian 50
Schaan, Kathy 50
Schacherer, Chris 9 , 47 , 60
Schafer, Lori 58
Scheibmeir, John 83
Schmiedl, Ryan 46
Schmitz, Amber 51 , 60
Schneider, Frank 35
Schneider, Mark 83
Schoeneberger, Jason 2
Schreiber-Gregory, Deanna 82
Schuelke, Matthew 27
Scocca, David 52 , 75
Sedlak, Doug 32
Seffrin, Robert 31
Selvakumar, Prashanthi 77
Sempel, Hans 29
Sethi, Saratendu 45
Seth, Vivek 28
Setty, Ashok 59
Shah, Monarch 13
Shankar, Charu 49
Shao, Lucheng 46
Sharda, Ramesh 38
Sharma, Priya 59
Sharma, Ruchi 78
Shen, Dongmin 19
Shenvi, Neeta 30 , 57
Shigaev, Victor 29
Shim, Kyong Jin 29
Shin, John 72
Shipp, Charlie 15
Shive, Wanda 59
Shu, Amos 16 , 76
Shubert, David 40
Siddiqi, Naeem 44
Silva, Alan 18
Simon, Jim 2
Sindhu, Neetha 25
Singh, Sarwanjeet 53
Skillman, Shawn 70
Skoglund, Jimmy 71
Slaughter, Susan 47
Sloan, Stephen 74
Smiley, Whitney 23 , 2
Smith, Casey 81
Smith, Helen 24 , 57
Smith, Kevin 57
Song, Wen 55
Soto, Michael 71
So, Ying 2
Sperrazza, Casey 82
Springborn, Robert 38
Srivastava, Anurag 24
Stephens, Rob 83
Stokes, Maura 1
Styll, Rick 41 , 70
Suau-Sanchez, Pere 74
Surratt, Lane 44
Svendsen, Erik 15
Szeto, Nora 7 , 54
T
Tabachneck, Arthur 39 , 56
Taitel, Michael 17
Tan, Hui Fen 14
Tao, Jill 3
Tavakoli, Abbas 15 , 65
Terry, Robert 27
Thiyagarajan, Sreedevi 28
Thompson, Casey 35
Thompson, David 26
Thompson, Stephanie 44 , 73
Thompson, Wayne 9 , 32
Thornton, Patrick 31
Thota, Srikanth 20
Tilanus, Erik 6 , 53
Timusk, Peter 14
Tin Seong, Kam 45
Tobias, Randy 3
Trahan, Shane 24
Traubenberg, Seth 50
Trivedi, Bharat 8
V
Valliant, Richard 5
Valverde, Roberto 29
Van Daele, Douglas 22
Van den Poel, Dirk 63
Vanderlooy, Stijn 5
Vandervort, Eric 37
Varney, Brian 39
Venkataramani, Meera 80
Villiers, Peter 31
Virgile, Robert 39 , 56
Virji, Shirmeen 25 , 28 , 53
Vitron, Christine 48 , 50
Voltes-Dorta, Augusto 74
Vralstad, Svein Erik 42
W
Wachsmuth, Jason 54
Wagner, James 5
Waller, Jennifer 1 , 48
Walters, Barbara 84
Wang, Cindy 32
Wang, Lihui 21
Wang, Stacy 50
Wang, Xiyun (Cheryl) 5 , 37
Wang, Ying 43
Wang, Yongyin 72
Warman, Nicholson (Nick) 36
Watson, Greg 75
Weber, Tom 42
Weiss, Michael 37
Weiss, Mitchell 69
Wells, Chip 9
Wexler, Jonathan 9
Whitehurst, Joe 39 , 56
Wicklin, Rick 49
Widel, Mario 11 , 75
Wilkins, Scott 46
Williams, Christianna 3 , 48
Williams, Simon 35 , 67
Wilson, Michael G. 2
Wolfe, Bryan 65
Wong, Eric 50 , 51
Wu, Yi-Fang 27
X
Xie, Fagen 43
Xu, Meili 18
Xu, Yan 75
Y
Yang, Dongsheng 24
Yang, Jason 55
Yao, Xue 27
Yiu, Sau 54
Yong, Chin Khian 22
Yuan, Yang 33
87
Z
Zakzeski, Audra 31
Zapata, Hector 82
Zaratsian, Dan 59
Zender, Cynthia 46 , 57
Zhang, Jiawen 43
Zhang, Ruiwen 44 , 61
Zhang, Sijian 76
Zhang, Xin 30 , 57
Zhang, Yu 43
Zhao, Beinan 50
Zhao, Zheng 59
zhao, juan 16
Zhong, Christina 75
Zink, Richard 75
Zuniga, Daniel 34
Zupko, William 52
88

Abstracts

Transcription

Similar documents

InterMail has a strong standing within digital and mail

Reducing Flight Operation Cost

SUGI 25: Web Based DSS Development

ROTTWEILER Update

OZ zorgverzekeringen Financiële resultaten 2001

Big Digital Data, Analytic Visualization, and the Opportunity of

sas gets on its soapbox over chemicals