With SAS 9.3, a SAS/GRAPH® software license is no longer required for ODS Graphics. The only SAS licenses required to use ODS Graphics are a Base SAS license and a license for the product that contains the procedure that you want to use. In addition, ODS Graphics is the default preference in the SAS windowing environment in SAS 9.3, and the default destination is now HTML with a new HTMLBLUE style. This paper is a great summary of the capabilities of ODS graphics.

The MCMC procedure, first released in SAS/STAT® 9.2, provides a flexible environment for fitting a wide range of Bayesian statistical models. Key enhancements in SAS/STAT 9.22 and 9.3 offer additional functionality and improved performance. The RANDOM statement provides a convenient way to specify linear and nonlinear random-effects models along with substantially improved performance. The MCMC procedure also supports multivariate distributions, such as the multivariate normal and inverse-Wishart distributions, and implements conjugate sampling methods when appropriate to improve sampling speed. This paper describes key enhancements in PROC MCMC and illustrates their use with examples.

Additional examples of the new features of PROC MCMC (different examples from those found in the SAS/STAT User's Guide) can be found on this website. These examples include Bayesian hierarchical Poisson models for overdispersed count data, the use of a multivariate prior for multiple linear regression, and a Bayesian multinomial model. 

If you have access to both SAS and JMP, you may be interested in a new interface to structural equation models (SEMs). Seamlessly integrating SAS and JMP software, SAS Structural Equation Modeling for JMP offers tools for creating and analyzing models. It provides an  interactive graphical interface for building path diagrams, translating the path diagrams into statements in the SAS/STAT® CALIS procedure, running the statements and displaying the results in JMP. You easily specify models with only observed variables (for example, regression and path analysis models) or models that have both observed and latent variables (for example, factor analysis and latent curve models). The intuitive model-building and analysis tools are suitable for both beginning and experienced structural equation modelers.

Rick Wicklin continues blogging about statistical programming, data analysis, and the SAS/IML language. Recently Wicklin reminded his readers of basic ideas behind constructing ROC curves and gave an example of how to write time-to-event simulations in SAS. He also shared some tips on finding the roots of multidimensional function by using Newton's method. The SAS blogging environment now has a new look and a new URL, so update your bookmarks to blogs.sas.com/content/iml to make sure that you don't miss any content. Even better, consider subscribing via an RSS feed.

Two keys to measuring and controlling the risk inherent in financial securities are (1) understanding the volatility of economic factors on which the value of the portfolio depends and (2) understanding how changes in those economic factors are related to each other. Recent progress in the mathematical technique of "copula" functions offers a powerful new approach to modeling dependencies among numerous risk factors. This paper explains how the new SAS/ETS® COPULA procedure performs copula modeling and shows examples of using copula models for risk management problems.

The finite population variance of a variable provides a measure of the amount of variation in the corresponding attribute of the study population's members, thus helping to describe the distribution of a study variable. Whether you are studying a population's income distribution in a socioeconomic study, rainfall distribution in a meteorological study, or scholastic aptitude test (SAT) scores of high school seniors, a small population variance is indicative of uniformity in the population while a large variance is indicative of a more diverse population. Another use for the population variance is to determine sample size. This Web example discusses how to estimate the variance of a variable in a finite population by using the SURVEYMEANS procedure.

Following is a list of tutorials scheduled for the fall SAS User Group regional conferences. Numerous other presentations by users and SAS staff are on tap for the analytical areas.

Introduction to Bayesian Analysis Using SAS® Software
(Maura Stokes)
NESUG 
Sept. 11-14 
Portland, ME

The Graph Template Language and the Statistical Graphics Procedures – An Example-Driven Introduction
(Warren Kuhfeld)
WUSS 
Oct. 12-14
San Francisco, CA

Introduction to Mixed Models
(Funda Gunes)
SCSUG 
Nov. 6-8
Fort Worth, TX

Data Simulation for Evaluating Statistical Methods in SAS®
(Rick Wicklin)
SESUG 
Oct. 23-25
Alexandria, VA

Multiple Comparison Methods in SAS/STAT® Software
(Randy Tobias)
PNWSUG 
Sept. 18, Seattle, WA
Sept. 20, Portland, OR

SAS will have a significant presence at the INFORMS (Institute for Operations Research and the Management Sciences) annual meeting November 13-16 in Charlotte, NC. INFORMS is a global professional organization that promotes the use of analytics in addressing organizational planning.

Keith Collins (Senior Vice President and Chief Technical Officer) will deliver a keynote address on Monday, Nov. 14, and SAS staff will present papers and workshops, chair sessions, and demo SAS software in the exhibition area.