 |
|
|
 |
Dear Reader,
Only 88 degrees outside, so it must be September in North Carolina. What a hot summer on the East Coast! I was one of the many statisticians who attended the Joint Statistical Meetings in Vancouver, Canada, in August. I kissed the cool Canadian ground when I got off the plane! I enjoyed the conference, hightailing it out of my committee meetings occasionally to listen to some great sessions. If you made it to Canada, I hope you had a great conference and took a gander at the SAS booth in the exhibition area.
The 2011 call for papers for SAS® Global Forum, April 4–7, 2011, is open, with all contributed paper abstracts due by Oct. 25. Think about submitting an abstract for a paper or poster, especially for the new Operations Research section. We’ll have several new statistical tutorials this year, so think also about joining us on Monday morning.
Model Selection with SAS/STAT® Software – Robert Cohen
Data Simulation for Evaluating Statistical Methods in SAS Software – Rick Wicklin
Multiple Comparison Methods in SAS/STAT Software – Randy Tobias
Creating Statistical Graphics in SAS Software – Warren Kuhfeld
Check the SAS Global Forum website later this fall for more information.
Read below for more information about our upcoming tutorials on the road. Also, find out about a new SAS blog, resources on support.sas.com, analyzing interval-censored data, capabilities for spatial data analysis, and more.
OK, you northern hemisphere people, time to rake some leaves!
Maura
R&D Research Director, Statistical Applications
P.S. All comments are welcome, particularly those about the types of information you want to see in upcoming editions of this newsletter.
|
|
SAS CEO Jim Goodnight Believes It’s an Exciting Time to Be a Statistician »
If you missed SAS CEO Jim Goodnight’s speech this year at JSM, you can still hear it courtesy of the ASA website. Goodnight talks about the growing need for statisticians and how they are becoming increasingly involved on the front lines of their organizations. Other available plenary addresses include the Deming lecture, in which Brent James discusses his role in health care reform, and the ASA Presidential Address, in which Sastry Pantula of North Carolina State University talks about the role of statistics in a data-centric world.
|
|
New Blog for Statistical Programming Using SAS/IML® Software »
Rick Wicklin, principal developer for SAS/IML software and SAS/IML Studio, recently completed a SAS Press book on the same topic, which will be available in October. Wicklin’s blog, with entries promised three times a week, is called The Do Loop, and it focuses on tips and techniques for writing efficient SAS/IML programs for data analysis, modeling, simulation, sampling, matrix computations, regression and data visualization. Check it out!
|
|
Web Resources for SAS® Statistical Graphics »
You might be one of the many SAS users just migrating from earlier versions of SAS software to SAS 9.2, including SAS/STAT 9.22 and SAS/ETS® 9.22. If so, you might be new to the wonders of SAS statistical graphics, which give you convenient access to more than 600 graphs in SAS/STAT and SAS/ETS software and provide you with the new SAS/GRAPH® SG procedures for generating graphs yourself. This new site gathers together various resources for statistical graphics: links to papers, books and documentation. We’ll be updating the site regularly.
|
|
Resources for SAS/STAT® 9.22 »
We also put together a page of links to papers and examples that highlight the new features of SAS/STAT 9.22, since many of you are dealing with learning about the new SAS/STAT 9.2 and SAS/STAT 9.22 enhancements at the same time. Included are examples on exact Poisson regression, model averaging, and power priors in the MCMC procedure.
|
|
Introducing SAS® Simulation Studio »
Discrete event simulation is used to model and investigate systems with complicated mathematical and logical relationships that make analytical solutions impossible – often featuring significant random elements and events. Simulation, one of the most widely used methods in operations research, captures these random elements to enable you to study system performance under varying conditions and configurations. It is employed in a wide variety of fields, including manufacturing, customer service and health care. SAS Simulation Studio is a graphical discrete event simulation application, available as part of SAS/OR® software and also as an add-on module for JMP® software (as SAS Simulation Studio for JMP).
|
|
Analyzing Interval-Censored Survival Data »
Survival data analysis is traditionally focused on analyzing lifetimes by using time that is measured to an event of interest, or the latest time available if the event did not occur during the observation period. Data measured in this way are called right-censored data. Many methods (nonparametric, semiparametric and fully parametric) have been developed over the years to deal with this type of data. But what methods are available if the event time is not directly observed and the event is known only to have occurred within some interval of time? Data measured in this way are called interval-censored survival data, and the use of SAS software to analyze this type of data is the focus of this paper.
|
|
Efficient Geostatistical Analysis with SAS/STAT® Spatial Procedures »
SAS/STAT 9.22 introduces exciting updates to the SAS suite of spatial procedures to augment, simplify, and streamline spatial analysis in one and two dimensions. All spatial procedures offer a wider variety of correlation models than before. Automated semivariogram modeling in the VARIOGRAM procedure reduces the time and effort required to select a suitable correlation structure. Item stores enable you to pass the semivariance information seamlessly to the KRIGE2D procedure for prediction or to the SIM2D procedure for simulation. Option-rich ODS Graphics plots give you the full picture for diagnosing your spatial data and visualizing your prediction and simulation results. The presentation demonstrates these new features and discusses key elements and underlying complexities of stochastic spatial analysis. With this knowledge, you can take full advantage of the latest features for your studies with spatial attributes.
|
|
Updated SAS® Enterprise Guide® 4.3 Provides a Slick SAS Programming Environment »
More than a few programmers at SAS headquarters are moving to the new SAS Enterprise Guide as their SAS programming environment of choice. While SAS Enterprise Guide is largely known as a point-and-click interface, its editor now offers syntax suggestions, autocompletion, syntax reference help, and a SAS program code formatter. This latest release became available in August; just talk to your SAS account rep if you’re interested.
|
|
Estimating Geometric Means with Survey Procedures »
Geometric means are widely used in a variety of scientific disciplines within the survey sampling community. Although the SAS/STAT survey procedures do not presently provide a direct method for estimating geometric means and their variances, it is still possible to estimate them. This example shows how to use the SURVEYMEANS procedure and a little SAS programming to estimate a geometric mean, its variance, and to compute confidence limits.
|
|
SIBS Day
This summer, the SAS Statistical R&D group hosted the SIBS (Summer Institute for Training in Biostatistics) students at North Carolina State University. The SIBS program introduces quantitatively oriented undergraduates to statistics as a career through a six-week session of coursework and field trips. After a tour of the SAS campus in Cary, NC, the students were treated to presentations on the life of a software developer and how to put statistical graphics into practice by Fang Chen and Bob Rodriguez, respectively. The visit concluded with a panel discussion of statistical careers at SAS with members of the testing, development, training and technical support staffs. SIBS is a great program, which has recently grown to include eight universities, so if you have young relatives or friends who might be interested, have them look into it! Schools begin accepting applications in December.
|
|
SAS Participates in Operations Research Conferences
In late 2010, SAS will have a significant presence at two prominent conferences: the INFORMS Annual Meeting in Austin, TX, and the Winter Simulation Conference in Baltimore. INFORMS (Institute for Operations Research and the Management Sciences) is the leading global operations research organization. On Nov. 7-10, INFORMS will hold its annual meeting at the Austin Convention Center and anticipates more than 4,000 attendees.
The Winter Simulation Conference is a smaller meeting focused on simulation of system behavior; it will be held Dec. 5-8 at the Marriott Waterfront Hotel in Baltimore. The Winter Simulation conference is jointly sponsored by several groups, including the Society for Modeling and Simulation International, the American Statistical Association, and the INFORMS Simulation Society.
|
|
SAS® Analytics Training Available »
A variety of SAS training is available through SAS Education, including a growing number of courses that are available over the Web. New SAS users might find these a valuable resource.
|
|
Plotting the fitted values from a random coefficients model »
After fitting a random coefficients model (also called a hierarchical linear model or HLM), you might want to graph the resulting fitted regression model for each subject. Specify the OUTP= option in the MODEL statement of PROC MIXED to create a data set to contain predicted values. Beginning in SAS 9.2, you can use PROC SGPLOT to plot the model.
|
|
Several SAS regional users group meetings will include presentations on the statistical areas and ODS Statistical Graphics. In addition, several two-hour tutorials will be presented:
An Introduction to SAS/IML® Studio for SAS/STAT Users by Rick Wicklin
WUSS Nov. 3-5, San Diego
Introduction to Logistic Regression by Maura Stokes
MWSUG Oct. 10-12, Milwaukee
The Graph Template Language and the Statistical Graphic Procedures: An Example-Drive Introduction by Warren Kuhfeld
NESUG Nov. 14-17
Introduction to Logistic Regression by Bob Derr
SCSUG Nov. 7-9, Austin, TX
Maura Stokes will be co-teaching a short course with Gary Koch at the Biopharmaceutical Applied Statistics Symposium (BASSXVII) at Hilton Head, SC, on Nov. 11-12.
|
|
Contact Us
Have a question, comment or suggestion? Want to see something covered in the next SAS Statistics and Operations News? Simply e-mail the editor. |
| |
| Tech Support | Privacy Statement |
|
