The latest release of SAS/STAT software brings you powerful techniques that will make a difference in your work, whether your data are massive, missing, or somewhere in the middle. New imputation software for survey data adds to an expansive array of methods in SAS/STAT for handling missing data, as does the production version of the GEE procedure, which provides the weighted generalized estimating equation approach for longitudinal studies with dropouts. An improved quadrature method in the GLIMMIX procedure means you can fit models not previously possible. The HPSPLIT procedure provides a rich set of methods for statistical modeling with classification and regression trees, including cross validation and graphical displays.

For a bird's-eye view of this new release, see the highlight video at http://support.sas.com/rnd/app/video/index.html#stat

The revised Chapter 3 of the SAS Simulation Studio 14.1 User's Guide contains updated information about the use of local and remote SAS servers in conjunction with SAS Simulation Studio. The default local SAS Workspace Server now launches automatically on demand. Credentialing for other local SAS Workspace Servers and Remote SAS Servers has been improved.

Since the financial crisis of 2008, banks and bank holding companies in the United States have faced increased regulation. One of the recent changes to these regulations is known as the Comprehensive Capital Analysis and Review (CCAR). At the core of these new regulations, specifically under the Dodd-Frank Wall Street Reform and Consumer Protection Act and the stress tests it mandates, are a series of "what-if" or "scenario analyses" requirements that involve a number of scenarios provided by the Federal Reserve. This paper proposes frequentist and Bayesian time series methods that solve this stress testing problem using a highly practical top-down approach. The paper focuses on the value of using univariate time series methods, as well as the methodology behind these models.

Balanced incomplete block designs (BIBDs) provide plans for certain types of experiments that are popular in many fields, including agricultural and marketing research. The SAS BIBD autocall macro %MktBIBD has been rewritten for the 14.1 release. Now, in addition to performing iterative searches, it has a BIBD catalog and construction methods based on Hadamard matrices. When iterations are not necessary, run time is often less than one second. When iterations are necessary, the default number of iterations has changed so that the macro will run faster than it did in previous releases. For b blocks, v treatments, and block size k, you can create a BIBD as follows:

     %mktbibd(b=20, v=16, k=4)

Other options are available as well.

In this video, learn about ROC curves from Principal Statistical Developer Bob Derr, who supports the LOGISTIC procedure. ROC curves can be useful for diagnostics and classification as well as model selection, and PROC LOGISTIC can not only output various ROC statistics but also fit different models and compare the resulting ROC curves by using the ROC and ROCCONTRAST statements. You can also use these statements to compare models produced with other statistical procedures.

Distinguished Research Statistician Developer Warren Kuhfeld describes how to customize every aspect of the ODS graphs that SAS procedures produce with three postings in the SAS blog  Graphically Speaking . Standard graph customization methods include template modification and SG annotation, which is usually used to annotate SGPLOT graphs. However, you can also use SG annotation to modify graphs that analytical procedures produce. Learn these techniques for modifying dynamic variables, annotating graphics from analytical procedures, and annotating multiple panels.

Learn about the basic principles of modeling panel data using SAS/ETS, particularly the PANEL procedure, from Roberto Gutierrez, senior research statistician developer.

Do you teach? Have you ever wanted to demonstrate the assumptions of various linear regression models? Learn how to use the SGPLOT procedure to visualize the response distributions of linear models.

Dealing with high-school-age children taking the SAT and ACT? Want to work with some test score data instead of worrying about their results? See how you can use SAS to visualize the distribution of recent ACT scores.

Occasionally the statistic of interest is a ratio of random variables. Two examples are the ED50 and relative potency, which are often of interest in bioassay studies. For such statistics, obtaining a confidence interval requires special methods. Two methods for obtaining confidence limits for a ratio of linear combinations of model parameters are presented. Zerbe (1978) showed how confidence limits based on Fieller's theorem can be obtained for any ratio of linear combinations of model parameters in a generalized linear model. The delta method is a general method that provides an approximate estimate of the variance of nonlinear functions of random variables. Hirschberg and Lye (2010) compare the two methods and discuss when they produce similar results.