Featured news from SAS Advanced Analytics

 

 ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ 
View as a web page
 
SAS | THE POWER TO KNOW
 
SAS Statistics and Operations Research News

My first step in putting this newsletter together was to get lunch. I found a seriously long line in the Café, with people waiting for Carolina (eastern) barbecue. The only meal that commands longer lines here is spaghetti, comfort food for the ages. Yes, we have some serious regional taste for barbecue going on and not just for people who were born here!

I ate my pre-newsletter meal while scrolling through a ASA Statistical Consulting Section thread on data science and how that relates to statistics.

What does barbecue have to do with data science and statistics and me? Well, let me tell you! As we work on SAS analytical software, we are constantly reminding developers to focus on a specific audience during the planning phases, be it a new procedure, new tasks in an interface, new user interface, or new programming paradigms. What does that audience do? How do they work? What’s their skill level? What helps them work better? Some of our audiences are just as regional, in some sense, as our different tastes in food.

A data scientist might focus on prediction with her modeling, whereas an epidemiologist might want to address specific hypotheses of interest. Interactions are of high interest to some modelers, a necessary nuisance to others, and to be avoided at all costs for a third group of modelers. How best to provide for each type of audience? Sometimes a general tool works for different audience, but sometimes you need specialized tools. Occasionally SAS may appear to offer similar capabilities in different products, but the context and terminology (say econometrics versus statistics) is so different that the customer wouldn’t be served well with a general tool.  

With statistics and machine learning, distinct audiences are less easily defined, with statisticians adopting unsupervised learning to contend with high dimensional data, and with data scientists expanding their statistical tools to build explanatory models. The border between statistics and machine learning is blurring, and the use of techniques is shifting to accommodate new kinds of data and different kinds of problems.   

Please keep in mind that you are our target audience, and we aim to serve all regions. Tastefully!  If we score, please let us know. And if we swung and miss (sorry, missing baseball during All- Star week in the United States but hey, go Portugal!), please let us know that, too! I just have a conversation with an instructor who thinks there is a growing demand for point-and-click access to survey data analysis. That’s exactly the type of feedback we want to hear from you --- about everything we do.

Many of us are on our way to JSM in early August and we’d love to talk with you there! We’ll be represented on the Exhibition Floor. It’s not too late to sign up.

And finally, for you people on the West Coast, the Western Users SAS Conference is being held on September 7-9 in San Francisco. It always has great analytical content.

Here’s to the rest of the summer.  

Maura Stokes 

Senior R&D Director, Statistical Applications

 

Technical Highlights

 

SAS/STAT Topic Areas

There’s a lot of stuff in SAS/STAT software. Like 10,000 pages of documentation worth. It can be overwhelming to determine what areas are covered and what procedures you should use. We introduced a Topics tab in the online documentation a few releases ago, and there are corresponding pages in our focus area that briefly describe the topic areas and the procedures that perform analyses for those areas. In the future, we’ll provide more linkage between the documentation and other web resources (videos come to mind), but for now, just use the URL support.sas.com/stattopics/ and see if this helps to navigate the SAS/STAT
landscape. 

 

paper icon

Spatial Dependence, Nonlinear Panel Models, and More New Features in SAS/ETS® 14.1

SAS/ETS 14.1 delivers a substantial number of new features to researchers who want to examine causality with observational data in addition to forecasting the future. This release adds count data models with spatial effects, new linear and nonlinear models for panel data, the X13 procedure for seasonal adjustment, and many more new features. This paper highlights the many enhancements to SAS/ETS software and demonstrates how these features can help your organization increase revenue and enhance productivity.

 

SAS/STAT Users Still Moving to the SAS 9.4 Platform? 

Aren’t we all! We featured this handout last time, but it’s timeless!

If you are moving up to SAS 9.4 and would like to catch up on the recent SAS/STAT
releases on that platform, this handout is for you! Get an overview of our new additions in missing data analysis, modern survival data analysis, statistical modeling, spatial point pattern analysis, Bayesian analysis, item response analysis, and classification and regression trees, along with many performance enhancements. There’s truly something in there for everyone. And if you aren’t currently on the move, feel free to use this handout however it helps you get into the passing lane! 

 

paper icon

Using SAS® Simulation Studio to Test and Validate SAS/OR® Optimization Models

In optimization modeling, you almost always need to make some simplifying assumptions about the details of the system you are modeling. These assumptions are especially important when the system includes random variation—for example, in the arrivals of individuals, their distinguishing characteristics, or the time needed to complete certain tasks. A common approach holds each random element at some nominal value (such as the mean of its observed values) and proceeds with the optimization. You can do better. In order to test an optimization model and its underlying assumptions, you can build a simulation model of the system that uses the optimal solution as an input and simulates the system’s detailed behavior. The simulation model helps determine how well the optimal solution holds up when randomness and perhaps other logical complexities (which the optimization model might have ignored, summarized, or modeled only approximately) are accounted for. Simulation might confirm the optimization results or highlight areas of concern in the optimization model. This paper describes cases in which you can use simulation and optimization together in this manner and discusses the positive implications of this complementary analytic approach.

 

paper icon

Tips and Strategies for Mixed Modeling with SAS/STAT Procedures

Inherently, mixed modeling with SAS/STAT procedures (such as GLIMMIX, MIXED, and NLMIXED) is computationally intensive. Therefore, considerable memory and CPU time can be required. The default algorithms in these procedures might fail to converge for some data sets and models. This encore presentation of a paper from SAS® Global Forum 2012 provides recommendations for circumventing memory problems and reducing execution times for your mixed-modeling analyses. This paper also shows how the new HPMIXED procedure can be beneficial for certain situations, as with large sparse mixed models. Finally, the discussion focuses on the best way to interpret and address common notes, warnings, and error messages that can occur with the estimation of mixed models in SAS® software.

 

The Do Loop

Distinguished Research Statistician Rick Wicklin shows you how to benefit from the EFFECTPLOT statement, which produces plots that show the predicted response as a function of certain covariates while other covariates are held constant. Available in several SAS/STAT regression procedures, it’s also provided in the PLM procedure. Learn how to use this underutilized tool.

In addition, this posting provides you with 10 tips when you run an optimization with SAS/IML®. This is one to cut out and tape to your desk.  

And finally, staying with the foodie motif of this newsletter, Wicklin shows you how to create lasagna plots in SAS when the spaghetti plot doesn’t quite cut it.

 

paper icon

Statistical Model Building for Large, Complex Data: Five New Directions in SAS/STAT Software

Inherently, mixed modeling with SAS/STAT procedures (such as GLIMMIX, MIXED, and NLMIXED) is computationally intensive. Therefore, considerable memory and CPU time can be required. The default algorithms in these procedures might fail to converge for some data sets and models. This encore presentation of a paper from SAS® Global Forum 2012 provides recommendations for circumventing memory problems and reducing execution times for your mixed-modeling analyses. This paper also shows how the new HPMIXED procedure can be beneficial for certain situations, as with large sparse mixed models. Finally, the discussion focuses on the best way to interpret and address common notes, warnings, and error messages that can occur with the estimation of mixed models in SAS® software.

 

paper icon

An Overview of ODS Statistical Graphics in SAS 9.4

ODS Statistical Graphics provides functionality for creating statistical graphics. ODS Graphics is available in the Base SAS®, SAS/STAT®, SAS/ETS®, and SAS/QC® products. More than 100 statistical procedures use this functionality, and they produce graphs as automatically as they produce tables. In addition, Base SAS procedures use ODS Graphics to produce plots for exploratory data analysis and for customized statistical displays. This paper presents the essential information that you need to get started with ODS Graphics in SAS® 9.4. When ODS Graphics is enabled, graphs and tables are integrated together in your ODS output destination. ODS Graphics produces graphs in standard image file formats, and the consistent appearance and individual layout of these graphs are controlled by ODS styles and templates, respectively. With some understanding of the underlying Graph Template Language, you can modify the default templates to make changes to graphs that are permanently in effect each time you run the procedure. Alternatively, you can make immediate changes by using the ODS Graphics Editor, whose point-and-click interface enables you to customize titles, annotate points, and make other enhancements.

 

Tech Support Points Out

 
Tech Support Points Out icon

Understanding Splines in the EFFECT Statement

A spline is a piecewise polynomial function in which the individual polynomials are of the same degree and which connect smoothly at join points whose abscissa values, referred to as knots, are predefined. You can use splines to obtain a more flexible fit to the data than can be obtained with a regular linear regression model.

Spline effects can be defined using the EFFECT statement that is available in several procedures. See the discussion of splines in the documentation of the EFFECT statement in the “Shared Concepts and Topics” chapter of the SAS/STAT User’s Guide. The example that follows in this SAS Usage Note illustrates fitting a model containing a spline effect in PROC GLIMMIX. It discusses the spline basis output, the interpretation of the output, how to use the spline model to make predictions, and how to use the LSMEANS and ESTIMATE statements to compute quantities of interest.

You can also score models containing spline effects. 

 

Talks and Tutorials

 
JSM logo

JSM 2016

The following Computer Technology Workshops are being taught at JSM 2016 in Chicago this summer.

Advanced ODS Graphics Examples in SAS by Warren Kuhfeld

Small Area Estimation Using SAS Software by Pushpal Mukhopadhyay

Weighted GEE Analysis Using SAS/STAT Software by Michael Lamm

Current Methods in Survival Analysis Using SAS/STAT Software by Changbin Guo

 

Resources

 

SAS® Statistics and Operations Research News

Don't miss important updates from SAS! Please add sas.com as a domain in your safe sender list.
Want to get more out of your relationship with SAS? Update your SAS profile
.

About this e-mail:
If you do not wish to receive future SAS® Statistics and Operations Research News editions, please unsubscribe.
SAS places great value in fair information practices and in connection with the management of our contact database, we would like to remind you that SAS Institute Inc. may use your personal contact details for marketing purposes, as stated in its privacy policy. To contact SAS via postal mail: SAS, SAS Campus Drive, Cary NC 27513 USA. ATTN: Legal Division/Privacy Manager.

SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration. Other brand and product names are trademarks of their respective companies. Copyright © SAS Institute Inc. All rights reserved.