I wanted to make sure you saw the 12.1 release number! That's right, the release-numbering scheme for the analytical products has changed with this release. The number is no longer the same as the required Base software number. We're independent of the Base mother ship and so we needed a different release-numbering scheme. So while Base SAS® remains at the 9.3 level, and the latest version is 9.3m2 (for the latest maintenance release), SAS/STAT is updated from SAS/STAT 9.3 to SAS/STAT 12.1. The next major release of SAS/STAT will be SAS/STAT 13.1.

Handouts for SAS/STAT 12.1 and SAS/ETS 12.1 are now available, and overview videos for SAS/STAT 12.1 and SAS/ETS 12.1 are also available. The Statistics and Operations Research focus area on support.sas.com also has lots of information on the 12.1 releases.

This JMP add-in uses SAS/ETS procedures to perform its computations. Version 2.1 enables users to estimate and refine autoregressive or heteroscedastic error models, unobserved component models, panel analysis models, count data models, and models of the probability of continuous valued events using the dynamic, interactive environment of JMP. Learn how to get started with this video.

The upcoming INFORMS meeting, “Informatics Rising,” is being held October 14–17 in Phoenix, AZ.  INFORMS, the Institute for Operations Research and the Management Sciences, is an international organization and one of the leading proponents of analytics. A number of SAS staff members will attend, present papers, and participate in panel discussions. The SAS Operations Research R&D staff will host a pre-conference workshop, “Solving Business Problems with SAS Analytics and OPTMODEL,” on Saturday, October 13. SAS will also have a presence on the exhibition floor, so come by if you are at the conference this year.

We recently said goodbye to five graduate summer students who were the winners of our annual Analytical Summer Fellow competition. The program was developed more than five years ago to expose interested students to the world of commercial statistical software research and development. This year the program was expanded to five fellowships, including new ones for data mining, operations research, and econometrics.

The program matches the students' skills and interests with specialized analytical needs. Each fellow is paired with a senior developer, who directs a summer project that can range from research to testing to programming. Fellows receive a stipend enabling them to move to Cary and also earn a summer student salary for 12 weeks. This summer's fellows were:

• Jared Erickson (Northwestern University)
• Jiangtao Gou (Northwestern University)
• Xiaoyan Sun (Emory University)
• Guannan Wang (University of Georgia)
• Zhenyu Zhao (Northwestern University)

The PLM procedure performs postfitting statistical analyses on model information saved from a previous analysis with a linear modeling procedure. This means that you can perform tasks such as testing hypotheses, computing confidence intervals, producing prediction plots, and scoring a new data set separately from fitting the original model. The PLM procedure can be called multiple times, thus saving you the model fitting expenditure and enabling you to perform these tasks when you may not still have access to the original data. This video provides you with an overview of PROC PLM.

SAS/OR 12.1 debuts a broad range of new capabilities and enhanced features in operations research. This paper surveys these additions, emphasizing their benefits and their application to business problems, and especially to large business problems. Optimization improvements include faster problem generation and support for additional specialized solution methods in the OPTMODEL procedure. SAS/OR 12.1 can also exploit multicore computing in an updated version of the nonlinear programming multistart method and in underlying methods for the LP, QP, and NLP solvers. Other updates include an expanded set of constraint classes for the CLP procedure and enhanced SAS® Simulation Studio support for large models and large experiments (many factors and responses).

If you've ever wanted to get your hands on the actual values being displayed by an ODS graphic, Rick Wicklin tells you how in his blog, The DO Loop .

You may also be interested in returning multiple values from a SAS/IML® function: http://blogs.sas.com/content/iml/2012/08/20/how-to-return-multiple-values-from-a-sasiml-function/

SAS® High-Performance Forecasting offers an innovative process for automatically combining forecasts. Forecast combination, also called ensemble forecasting, is the subject of many academic papers in statistical and forecasting journals; it is a known technique for improving forecast accuracy and reducing variability of the resulting forecasts. This paper describes this new capability and includes examples that demonstrate the use and benefits of this new forecast combination process.

These regional conferences feature statistical tutorials and talks, plus the great opportunity to engage with SAS users in your area.

PROC SGPLOT does not have a QQPLOT statement like the one available in PROC UNIVARIATE, but you can use the SCATTER statement to create normal quantile-quantile plots by first computing the normal quantiles of your data.