Five new features of SAS 9.3 for statistical programmers


Welcome, SAS 9.3! I've already blogged about some interface and graphical changes that everyone should know about. Now I'll put on my statistical hat and mention a few 9.3 features that excite me, personally, as a data analyst and a statistical programmer:

  1. As a statistician, I am keen to try the new FMM procedure, which fits finite mixture models. This regression procedure in SAS/STAT software should be useful when the response variable is a finite mixture of univariate distributions. You can also use it to model univariate data as a mixture of, say, three normal distributions. Sweet!
  2. As someone who runs a lot of simulations, I am eager to try the COPULA procedure in SAS/ETS software. The COPULA procedure enables you to simulate data from a multivariate distribution that has a given covariance structure. I've written a copula algorithm in SAS/IML, so I am looking forward to writing a "couple 'a" blog posts on comparing the two approaches. Plus, it's fun to say `kä-pyə-lə. Copula! Copula! Copula!
  3. As a statistical programmer, I am stoked about the new features in SAS/IML 9.3, such as the ability to call SAS procedures from within a SAS/IML program. (Technically, this feature has been available since SAS/IML 9.22, which shipped November, 2010.) In the fall, I intend to blog about the new SAS/IML features, but if you can't wait, you can browse the What's New in SAS/IML 9.3 document.
  4. I use the SAS/IML interface to the R language, and SAS/IML 9.3 enables me to call the most recent version of R. This corrects a problem that occurred when the R directory structure changed between R 2.11 and 2.12.
  5. Lastly, I use the SAS online documentation a lot while doing my SAS programming, so I was pleased to see several changes to the organization of the documentation.
    • The large SAS Language Reference: Dictionary has been divided into seven documents. That should save me a few clicks.
    • The main SAS Product Documentation page has been redesigned. If you know what you're looking for and which book it is in, click Find documentation for a specific product. Otherwise, there are new ways to search the documentation.
    • The documentation for each procedure now has "tabs" that enable you to navigate directly to sections or subsections of the documentation. For example, the image at the end of this post shows the documentation page for PROC COPULA, and a drop-down menu that enables you to navigate within the "Details" section.

What is your favorite new statistical or SAS programming feature? Browse the What's New in SAS 9.3 document and let me know.


About Author

Rick Wicklin

Distinguished Researcher in Computational Statistics

Rick Wicklin, PhD, is a distinguished researcher in computational statistics at SAS and is a principal developer of SAS/IML software. His areas of expertise include computational statistics, simulation, statistical graphics, and modern methods in statistical data analysis. Rick is author of the books Statistical Programming with SAS/IML Software and Simulating Data with SAS.


  1. Charlie Huang on

    I am very interested in your future blog post comparing the new Copula procedure and the IML program.

  2. Pingback: ODS Graphics: new features and bundling in SAS 9.3 - The SAS Dummy

  3. Pingback: Modeling Finite Mixtures with the FMM Procedure - The DO Loop

      • Onyemachi Chris Uche on

        I am glad for your response, however, I am still worried if the SAS discriminant procedures could solve my problem in DISCRETE DISCRIMINANT analysis. And please direct me to obtain a copy of the software.

          • Onyemachi Chris Uche on

            I find your response very helpful, having referred to Butler (1998).
            Butler indicates that SAS procedures for nonparametric discriminant
            analysis are introduced and analyzed for their use with discrete data, using
            a combination of Base SAS and SAS/IML. How can I obtain this resource for
            a student research work that I am carrying out?

Leave A Reply

Back to Top