 |
|
|
SAS® Statistics and Operations Research News
|
|
|
|
 |
Dear Readers,
Spring has landed in North Carolina. Plants are blooming, and, with any luck, my flower beds will be covered with new mulch by the time you read this newsletter! The 2009 SAS Global Forum has come and gone; however, I remain grateful that the SAS IT folks bailed me out with a spare power cord right before my presentation. Apparently I need to move my packing list to my new smart phone!
This is the third newsletter, and it’s great to see the audience growing. This edition focuses on the new features available with SAS® 9.2, which many sites are now receiving. Our interview is with Bob Derr, Senior Research Statistician Developer. Also learn about the new discussion forums and the winner of the 2009 SAS Statistical Summer Fellowship.
Enjoy!
Maura
P.S. All comments welcome, particularly about the types of information you want to see in upcoming editions.
|
|
An Interview with Bob Derr, Senior Research Statistician Developer »
Bob Derr is a Senior Research Statistician Developer with the Linear Models Department at SAS. He began working here in 1994 while he was a student in the Statistics Department at the University of North Carolina at Chapel Hill, and he accepted a full-time position in 1998. Before his move to North Carolina, Bob had received a master’s degree in mathematics with a concentration in statistics at Indiana University. He then worked as a statistician at the Bureau of Labor Statistics in Washington, DC, for three years.
Bob supports the CATMOD, LOGISTIC and MULTTEST procedures, in addition to providing leadership in the ODS Graphics development project. He implemented newer techniques, such as exact logistic regression, in PROC LOGISTIC, and he also programmed the adaptive p-value adjustments in PROC MULTTEST.
Read this interview about the life of a SAS/STAT® developer.
|
|
New Discussion Forums »
We’ve recently added two new discussion forums: “SAS Statistical Procedures” and “SAS/IML® and SAS/IML Studio.” These forums are intended primarily for user-to-user discussions and are not intended to replace the role of SAS Technical Support. However, we hope they will provide a rich source of information for you and serve as a channel for sharing best practices with other users. Please check them out!
|
|
10 Great Reasons for a Statistician to Upgrade to SAS® 9.2 »
With the full release of SAS 9.2 now available (an earlier release provided just the base product as well as the analytical products), many sites are making the decision about whether to install it. Many users approached us at SAS Global Forum 2009, asking for help with convincing their management that installing SAS 9.2 would be a smart move. So we’ve produced a list of 10 great reasons for statisticians to install SAS/STAT 9.2. See what you think!
|
|
Getting Started Resources for SAS/STAT® 9.2 »
We’ve organized a collection of useful resources for customers who are beginning to work with SAS 9.2. This Web page provides links to related information on the support.sas.com site, including past SAS Global Forum papers, particularly useful chapters in the documentation and examples. We will probably modify this page as we develop more resources, so please let us know if this is useful.
|
|
Statistical Summer Fellow Awarded to Bingqing Zhou
For the past four years, we have held a competition for the SAS Summer Statistical Fellow. The applicants are graduate students across the United States who have completed at least two years of graduate school and are interested in statistical computing as a career. The winner lands a summer internship at SAS in addition to a stipend. The internship involves a research project with a developer in the student’s area of interest. This year, the winner is Bingqing Zhou of the Biostatistics Department in the School of Public Health at the University of North Carolina, who bested a field of over 40 applicants for this award. Bingqing recently won a student award at the past ENAR conference. She will be starting her internship this summer, working with us in the field of survival analysis.
Note that the call for applications for this program goes out in December of each year.
|
|
New Bayesian Modeling Tool »
Besides providing Bayesian analyses in the GENMOD, LIFEREG and PHREG procedures that were first introduced as experimental downloads for SAS 9.1.3, SAS/STAT 9.2 includes a tool for general Bayesian computations. The MCMC procedure enables you to fit a wide range of complex Bayesian statistical models. It uses the Markov chain Monte Carlo (MCMC) algorithm to draw samples from an arbitrary posterior distribution, which is defined by the prior distributions for the parameters and the likelihood function you specify.
|
|
Structural Analysis of Time Series »
Structural models provide regression-like decomposition of the response series into components (such as trend, seasonal or other periodic components) and linear and nonlinear regression effects. Apart from the series forecasts, structural modeling provides estimates of these unobserved components; these estimates are very useful in practical decision making.
The UCM procedure in SAS/ETS® software provides analysis of time series data by using structural models. In SAS 9.2, the UCM procedure contains several new features: approximation of long and complex seasonal patterns by using splines and trigonometric polynomials, incorporation of linear and nonlinear regression effects with time-varying coefficients, detection of structural change, and additional ODS graphics (including a variety of diagnostic plots).
|
|
Using ODS Statistical Graphics in SAS/ETS® Software »
Graphical visualization of statistical results is increasingly important in all areas of data analysis, and SAS/ETS procedures now provide graphical output for many kinds of econometric and time series models. The correct use of these features helps you to diagnose data patterns and choose appropriate models. In this paper, several SAS/ETS procedures, including the AUTOREG, PANEL, UCM and SIMILARITY procedures, are used to demonstrate a graphical diagnostic tool for effective modeling of cross-sectional, time-series, and panel data. The emphasis of the paper is on correct model selection and effective use of SAS/ETS software.
|
|
Optimization in SAS® Software »
Recently, there has been a lot of focus on the value of optimization in a variety of application areas. While optimization methods have been available in SAS for years, primarily in the SAS/OR® solution, they have become far more powerful and accessible in recent releases. In this recent SAS Global Forum paper, Ed Hughes and Trevor Kearney provide a nice overview of optimization in addition to a survey of optimization techniques. They then describe the latest additions to the optimization capabilities in SAS/OR software, with a focus on the OPTMODEL procedure.
|
|
Get More Value from ODS Statistical Graphics »
Some of you have used ODS Statistical Graphics in SAS 9.2. You may be familiar with the SG procedures, which provide common graphs used in data analysis. However, you may not know that you can combine the statements and options in these procedures to create specialized graphs in creative ways. Dan Heath, one of the SG developers, describes several techniques that you might find very useful.
|
|
Tech Support Points Out – Exact CMH Tests »
The CMH option in the TABLES statement of PROC FREQ is often used to obtain a test of the association between the row and column variables adjusted for the strata. Only asymptotic CMH tests, which require large overall sample sizes, are available in PROC FREQ. For stratified 2 × 2 tables, an exact test can be obtained using the STRATA statement and PERM= option in the MULTTEST procedure. The LOGISTIC procedure can also provide an exact test when the EXACT and STRATA statements are specified. This note provides examples for performing this test, including examples for the ordinal case.
|
|
If you live in Michigan, note that the Michigan SAS Users Group is holding a one-day SAS conference May 12. I’ll be speaking on Bayesian analysis and the new features of SAS/STAT 9.2.
PharmaSUG is holding its annual conference in Portland, OR, May 31 to June 4, and yes, I’m speaking at that one, too! I’m giving a tutorial on logistic regression, and I’m looking forward to meeting with conference attendees in the demo room as well.
The Joint Statistical Meetings are scheduled for Aug. 1-6 in Washington, DC. SAS R&D will give four tutorials as part of the Computer Technology Workshops. These tutorials provide methodological overviews as well as illustrate applications with SAS software. Registration began May 1.
Introduction to Bayesian Analysis Using SAS Software: Fang Chen
Using Replication Methods to Analyze Survey Data in SAS Software: Tony An
Introduction to the MCMC Procedure in SAS/STAT Software: Fang Chen
Group Sequential Analysis Using SAS Software: Yang Yuan
We’ll also teach various tutorials at the SAS regional users group meetings this fall; more details on that in the August newsletter.
|
|
Contact Us
Have a question, comment or suggestion? Want to see something covered in the next SAS Statistics and Operations News? Simply e-mail the editor. |
| |
SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration. Other brand and product names are trademarks of their respective companies. Copyright © 2009 SAS Institute Inc. All rights reserved
Tech Support | Privacy Statement |
|
