Online Program

Keynote Presentation | Concurrent Sessions | Poster Sessions
Short Courses (full day) | Short Courses (half day) | Tutorials | Practical Computing Expos | Closing General Session with Refreshments

Last Name:

Abstract Keyword:



Viewing Tutorials onlyView Full Program
Saturday, February 22
T1 Creating Statistical Graphics with ODS in SAS® Software
Sat, Feb 22, 1:30 PM - 3:30 PM
Bayshore V
Instructor(s): Warren F. Kuhfeld, SAS

Download Handouts
SAS 9.2 provides ODS Graphics, new functionality used by statistical procedures to create statistical graphics as automatically as they create tables. ODS Graphics is also used by new Base SAS procedures designed for graphical exploration of data. This tutorial is intended for statistical users and covers the use of ODS Graphics in statistical analysis.

You will learn how to:
- Request graphs created by statistical procedures
- Use the new SGPLOT, SGPANEL, SGSCATTER, and SGRENDER procedures to create customized graphs
- Access and manage your graphs for inclusion in web pages, papers, and presentations
- Modify graph styles
- Make immediate changes to your graphs using a point-and-click editor
- Make permanent changes to your graphs with template changes
- Specify other options related to ODS Graphics

Outline & Objectives

1. Introduction:

a. The basics of ODS Graphics

i. Enable ODS Graphics

ii. Procedures that support ODS Graphics

b. One step beyond the basics

i. Destinations

ii. Optional graphs

iii. Change graph style

iv. Editable graphs

v. Commonly used options

2. Graph and Style Template Languages

a. Templates and item stores

i. Template source

ii. SAS libraries

iii. ODS path

iv. Clean up

b. Graph template modification

i. Titles, ticks, axis labels, grids

ii. Adding date and project information

iii. Axis labels

iv. Graph data object

v. Dynamic variables

vi. Survival plot

c. Style template modification

3. The SG procedures and the GTL

a. Graph Template Language (GTL)

b. PROC SGPLOT - scatter plot, fit plot, histogram and density plots, bar chart, box plot, text insets, unicode, series plot and drop lines

c. PROC SGSCATTER - scatter plot matrix, panel of scatter plots

d. PROC SGPANEL - Data panel

e. Residual Panel (with GTL)

4. Conclusions

About the Instructor

Warren F. Kuhfeld is director of the SAS Advanced Regression Models R&D group. He received his Ph.D. in psychometrics from UNC Chapel Hill in 1985 and joined SAS in 1987. He has used SAS since 1979 and has developed SAS procedures since 1984. Warren wrote the SAS/STAT documentation chapters, “Using the Output Delivery System,” “Statistical Graphics Using ODS,” “ODS Graphics Template Modification,” and the SAS Press book Statistical Graphics in SAS: An Introduction to the Graph Template Language and the Statistical Graphics Procedures. Warren currently develops eleven SAS/STAT procedures and over twenty SAS macros for experimental design.

Relevance to Conference Goals

This tutorial addresses Theme 4: Software and Graphics.
Participants will learn how to create, manage, and customize a variety of graphs. Modern statistical graphics enable statisticians to better communicate statistical information. Knowledge of ODS Graphics is an indispensable skill for statisticians who use SAS to manage and analyze their data.

T3 Model Selection for Linear Models with SAS/STAT® Software
Sat, Feb 22, 1:30 PM - 3:30 PM
Bayshore VII
Instructor(s): Funda Gunes, Research Statistician at SAS Institute

Download Handouts
When you are faced with a predictive modeling problem that has many possible predictor effects, a natural question is, ”What subset of the effects provides the best model for the data?” This workshop explains how you can address this question with model selection methods in SAS/STAT software. The workshop also explores the practical pitfalls of model selection. The workshop focuses on the GLMSELECT procedure and shows how it can be used to mitigate the intrinsic difficulties of model selection. You will learn how to use model selection diagnostics, including graphics, for detecting problems; use of validation data to detect and prevent under-fitting and over-fitting; modern penalty-based methods, including LASSO and adaptive LASSO, as alternatives to traditional methods such as stepwise selection; and bootstrap-based model averaging to reduce selection bias and improve predictive performance. This workshop requires an understanding of basic regression techniques.

Outline & Objectives

Traditional model selection methods,
Customizing the selection process,
Penalized regression methods,
Model averaging,
Selection for nonparametric models with spline effects

The objective of this course is to teach the goals of model selection and difficulties of implementing it for a real data. The participants will learn how to mitigate these difficulties using extensive capabilities of the GLMSELECT procedure of SAS/STAT software.

About the Instructor

I am a Research Statistician in the Statistical Applications Department at SAS. I completed my PhD in statistics from North Carolina State University with focus in model selection methods.

As a research statistician in statistical R&D at SAS, I often give expository talks and two-hour tutorials at JSM and ENAR on a variety of topics, including mixed models, model selection, and Bayesian statistics. These presentations emphasize basic concepts, and they introduce applied statisticians to new methodology with relevant examples.

Relevance to Conference Goals

I attended the first CSP in 2012 and was impressed by the goals of the conference. I especially enjoyed meeting statistical practitioners and learning how they use statistics in their work. Based on that experience, I think the CSP audience is a perfect fit for the content and level of the presentations that I give, and I would love to participate.

T4 Learning and Improving Skills to become an Effective Statistical Collaborator
Sat, Feb 22, 1:30 PM - 3:30 PM
Palma Ceia II
Instructor(s): Eric Vance, Virginia Tech

Download Handouts
This fast-paced tutorial will get you up to speed on how to effectively collaborate with nonstatisticians to solve real-world problems and implement solutions. Based on Vance’s methods used to train more than 130 graduate students to become effective statistical collaborators for LISA (Virginia Tech’s Laboratory for Interdisciplinary Statistical Analysis), this tutorial will introduce you to the POWER process for structuring efficient meetings, lead you through role plays to practice what you have learned, and show you how to systematically improve your statistical collaboration skills through the recording and analysis of video.

Outline & Objectives

Tutorial participants will have read about the POWER process before arriving at the conference and will work in groups at the beginning of the tutorial to apply their understanding of the readings to answer questions and solve problems in the context of statistical consultations and collaborations. Dr. Vance will briefly lecture on any concepts of the POWER process the participants had trouble understanding or applying. At the end of this section of the tutorial, participants will solidly understand the five steps in the POWER process that lead to effective statistical collaborations.

The second section of the tutorial will consist of role plays in which the participants practice the key concepts of the POWER process. Alternating between the roles of client, statistical collaborator, and observer—while giving feedback to their peers—the participants will practice the skills they will use to unlock their collaborative potential.

The last section will show participants how to video record themselves during meetings with clients and, more importantly, how to analyze that video to systematically improve their statistical collaboration skills upon their return to their organizations.

About the Instructor

Dr. Eric Vance is a statistician and data scientist in the Department of Statistics at Virginia Tech. He is the Director of LISA (Laboratory for Interdisciplinary Statistical Analysis). LISA’s mission is to train statisticians to become statistical collaborators and promote the value of statistical thinking in all phases of scientific research by helping researchers design experiments; collect, analyze, and plot data; run statistical software; interpret results; and communicate statistical concepts to non-statisticians. From 2008-2012, LISA has helped solve research problems on 1,629 collaborative projects in 76 departments at Virginia Tech, helped 1,262 visitors during Walk-in Consulting, and taught the application of statistics to 2,223 attendees at LISA Short Courses for a total of 5,114 LISA clients in its first five years in existence.

Relevance to Conference Goals

This tutorial will help the participants develop statistical collaboration skills that will improve their personal effectiveness as statisticians. Participants will learn best practices in statistical consulting and collaboration that will enhance their organizational impact and potentially lead to career development and advancement. Participants will return to their jobs with new ideas, techniques, and strategies to improve their ability to communicate and collaborate effectively, resulting in a greater impact on their organizations.