 |
|
|
 |
Dear Reader,
I have World Cup soccer fever, so it seems like a great time to thank our readers from all over the world for subscribing to this newsletter. Our statistical staff has a definite international flavor, with recent hires hailing from Ethiopia, Turkey and China. People from India, Greece, Germany, China and South Africa grace my own hall, making my native state of New Hampshire seem a bit less exotic here in North Carolina!
Last issue I talked about the winter’s snow; today it’s 95 degrees out there. Whoa!
Well, the news of the season is that we have new releases out for the SAS/STAT®, SAS/ETS® and SAS/OR® products. We are determined to get analytical software updates out to our customers faster, and the 9.22 releases represent our efforts to do so. Available in conjunction with the third maintenance release of Base SAS® 9.2, major new enhancements are included in these products. See below for more information about these releases.
Many of us will head for the Joint Statistical Meetings in Vancouver, British Columbia, later this summer. Please stop by the SAS booth and say hello if you attend the meetings. The Talks and Tutorials section below describes some of the SAS courses that might be of interest to you at JSM this year.
Have a great summer, and I hope your favorite team (and mine) is still playing in the World Cup!
Maura
R&D Research Director, Statistical Applications
P.S. All comments are welcome, particularly those about the types of information you want to see in upcoming editions of this newsletter.
|
|
New Releases: SAS/STAT®, SAS/ETS® and SAS/OR® 9.22 »
Take a look at the new SAS/STAT, SAS/ETS and SAS/OR releases that came out in conjunction with the third maintenance release of Base SAS 9.2, available right now. SAS/ETS includes a new procedure called PROC SEVERITY for fitting models for statistical distribution of the magnitude of events, SAS/OR provides richer support for customized solutions to optimization problems in the OPTMODEL procedure, and SAS/STAT introduces two new procedures and numerous enhancements.
|
|
Overview of SAS/STAT® 9.22 »
As I was saying, SAS/STAT 9.22 introduces a number of updates. The new SURVEYPHREG procedure performs survival analysis of sample survey data, and the CALIS procedure has been totally revamped to include capabilities that were previously included in the TCALIS procedure of SAS/STAT 9.2. Exact Poisson regression, zero-inflated negative binomial regression, and seamless analysis and prediction of spatial processes are also included. For more information, see The Next Generation: SAS/STAT® 9.22.
|
|
Postfitting Analyses for the Linear Modeling Procedures »
SAS/STAT software provides a variety of ways to fit linear models. You can perform tests concerning the model parameters with the use of statements such as ESTIMATE, TEST and LSMEANS. In SAS/STAT 9.22, additional postfitting statements have been added to many modeling procedures, including some newer statements such as the LSMESTIMATE statement and the SLICE statement. In addition, the new PLM procedure enables you to perform postfitting analysis on model information stored from the linear modeling procedures and to score a new data set with a saved model and produce model effect plots. These exciting enhancements will benefit most SAS/STAT users. This SAS Global Forum paper by Randy Tobias explains the new enhancements with a number of examples.
|
|
Nonlinear Optimization in SAS/OR® Software: Migrating from PROC NLP to PROC OPTMODEL »
The OPTMODEL procedure in SAS/OR is the flagship optimization procedure. It provides the major legacy algorithms previously supplied by the NLP procedure, but, in addition, provides access to new algorithms that can solve large-scale problems and deal with nonlinearity more robustly. In addition, PROC OPTMODEL includes programming capabilities that enable users to develop their own customized solution methods. It is an excellent replacement for the NLP procedure not only for operations research practitioners but also for statisticians, econometricians and data miners. Examples illustrate how to migrate from PROC NLP to PROC OPTMODEL and highlight the benefits for doing so.
|
|
Modeling the Severity of Random Events with the SAS/ETS® SEVERITY Procedure »
The new SEVERITY procedure fits probability distributions for the severity (magnitude) of random events. Important examples include events with negative impact (such as the distribution of losses claimed under insurance policies, the magnitude of damages caused by natural disasters and the severity of outbreaks of a disease) and events with positive impact (such as order sizes for products characterized by intermittent demand). The SEVERITY procedure can use any of eight different parametric families of probability distributions. If these eight built-in distributions are not sufficient for your problem, then you can extend the SEVERITY procedure to support any parametric family by specifying its CDF and PDF functions with SAS programming statements and the FCMP procedure.
|
|
Example of Using Bootstrap Replicate Weights with Survey Procedures »
Providing bootstrap replicate weights for the purpose of variance estimation is a common practice within the survey sampling community. Although the SAS/STAT survey procedures do not presently provide options specifically designed to accommodate bootstrap replicate weights, you can still incorporate bootstrap replicate weights. This example shows how and when to use bootstrap replicate weights with existing SAS/STAT survey procedures to perform variance estimation.
|
|
SAS® Global Forum 2010 Keynotes by David Dickey and Joseph Gardiner
If you attended SAS Global Forum in Seattle this year, you might have been fortunate enough to grab a seat for one or both outstanding keynote talks in the Statistics and Data Analysis section. Not everyone did, because the talks were standing room only and people had to be turned away (we’ll work on a bigger room for next year!). Fortunately, these presenters also wrote papers for the proceedings. Take a look at Ideas and Examples in Generalized Linear Mixed Models by David Dickey, PhD, of North Carolina State University for a review of the ideas behind the GLIMMIX procedure, and see Survival Analysis: Overview of Parametric, Nonparametric and Semiparametric Approaches and New Developments by Joseph Gardiner, PhD, of Michigan State University for an overview of the concepts of survival analysis.
|
|
Bob Rodriguez from SAS Elected to ASA President »
We’re thrilled to announce that Bob Rodriguez, Senior Director, Research and Development, has been elected as the president of the largest statistical organization in the world. Rodriguez will serve as the 107th president with his term that begins in January 2012. He has previously served as ASA Vice President and recently received the Founder’s Award for distinguished service to the association. One of his goals is to increase public awareness of the value and impact of the field of statistics. The ASA is the second-oldest continuously operating professional society in the United States.
|
|
SAS at JSM 2010
Many SAS staff members will attend the Joint Statistical Meetings in Vancouver, British Columbia, from July 31 to Aug. 5. SAS CEO Jim Goodnight will deliver the President’s Invited Address on Tuesday afternoon. SAS/STAT 9.22 and SAS/ETS 9.22 will be on display on the exhibition floor, in addition to a nearly complete user interface for structural equation modeling using the SAS/STAT CALIS procedure through JMP® software. At the SAS Publications booth, you can review new SAS Press books Statistical Programming Using SAS®
by A. John Bailer of Miami University and an upcoming book by SAS’ own Rick Wicklin, Statistical Programming Using SAS/IML® Software: An Introduction.
|
|
Analysis of Observational Health Care Data Using SAS® »
Edited by Douglas Faries, Andrew Leon, Josep Haro and Robert Obenchain, this new book from SAS Press provides a number of articles focused on the methodology for analyzing observational data using SAS procedures and macros. Such methods are increasingly important because observational data collected for purposes such as patient registries and health care claims databases are an important source of information for determining how treatments and policies work in practice. Statisticians, epidemiologists, medical researchers and health outcomes researchers might be interested in reviewing this volume.
|
|
SAS® OnDemand for Academics »
Announced at SAS Global Forum this year, this program is now offered at no cost to university professors and students beginning in fall 2010. It provides an online service for learning SAS software; you register and access SAS software via the Web and perform processing by connecting to a hosted server at SAS. Professors register their classes for the service, which takes advantage of SAS applications such as SAS® Enterprise Guide® and SAS® Enterprise Miner™.
|
|
SAS staff will present a continuing education course and several computer technology workshops at JSM this summer. Here’s the lineup:
Practical Bayesian Computation
Fang Chen
Modeling Loss Distributions Using SAS/ETS Software
Mahesh Joshi
Introduction to Structural Equation Modeling Using the CALIS Procedure in SAS/STAT Software
Yiu-Fai Yung
An Introduction to SAS/IML Studio for SAS/STAT Users
Rick Wicklin
|
|
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. |
| |
| Tech Support | Privacy Statement |
|
