Courses and Tutorials

Thursday, February 16, 2012

8:30 a.m.-5:30 p.m. Full Day Courses
8:30 a.m.-12:30 p.m. Half Day Courses

Saturday, February 18, 2012

1:30-3:30 p.m. Tutorials

Courses and Tutorials are ticketed events and have limited availability. If you intend to register for a short course or tutorial onsite, please check with the ASA office first to verify availability.

Thu, Feb 16

SC1: Analysis of Messy Data: Design and Analysis of Experiments Requiring Mixed Models

Thu, Feb 16, 8:30 AM - 5:30 PM
Atlantis B

Instructor(s): George Milliken, Kansas State University (retired)

Four basic design structures that are building blocks for complex designs will be presented visually and conceptually. The analyses of the four design structures will be described in such a way that one can determine the appropriate error terms and associated degrees of freedom. Three of the four design structures are mixed models. The concept of fixed and random effects will be discussed, as well as analysis of a basic mixed model. Several examples will be used to demonstrate the use of the basic design structures to extract an appropriate model for complex designs. Examples will include incomplete block designs, repeated measures designs, split-plot designs, strip-plot designs, and combinations of these designs. SAS’s PROC MIXED and PROC GLIMMIX will be used to demonstrate the construction of code from the results of developing models for complex designs.

 

SC2: Regression Modeling with Many Correlated Predictors: High-Dimensional Data Analysis in Practice

Thu, Feb 16, 8:30 AM - 5:30 PM
Atlantis A

Instructor(s): Tony Babinec, AB Analytics; Jay Magidson, Statistical Innovations Inc.

The availability of a vast amount of data in fields such as genomics, marketing research, and signal processing has led to recent advances in high-dimensional data analysis. It is now possible to develop reliable regression models, even when the number of predictors exceeds the number of cases. In this course, we will begin by reviewing problems and limitations with traditional linear and logistic regression. We will then introduce the two primary regularization approaches for analyzing such data—penalized regression and component methods—related software, and recent advances in feature selection. Our applications-oriented presentation provides insight into how the new approaches work in examples with both low- and high-dimensional data and an overview of the relevant theory, supplemented by supporting equations. We will use real and simulated data sets to illustrate the different methods. The material presented will be included in a forthcoming book on this topic by the presenter.

 

SC3: Introducing R for Statistical Analysis

Thu, Feb 16, 8:30 AM - 12:30 PM
Fantail

Instructor(s): Eric Nantz, Eli Lilly and Company

R is a powerful statistical and data analysis software package and has rapidly become the software of choice for understanding data and applying statistical methodologies. Unlike the traditional statistical analysis software, R is free and open-source. This course will introduce R and supplemental tools to enhance the user experience. We will then discuss how to import common data file formats into R for analysis. We will use example data sets to illustrate how to compute basic statistics, conduct inference procedures, fit linear models, create visualizations of the data, and perform data manipulations. Last, we will discuss best practices for writing R programs and additional resources for learning more about the capabilities of R. When finished, you will be able to use R for a comprehensive data analysis with basic data operations, statistical methods, and visualization. Attendees should have a basic understanding of descriptive statistics and statistical methods such as t-tests, linear regression, and chi-square test. Before the course, instructions for installing R and supplemental materials will be given to all participants. While not required, it is beneficial for attendees to install this software before the course.

 

SC4: Why Don’t They Get It?

Thu, Feb 16, 8:30 AM - 12:30 PM
Zander

Instructor(s): Bill Williams, Organizational Learning Consultant

Many of us make presentations to others as part of our work. These range in length and complexity of the situation, from presenting ideas to your supervisor, to presenting analysis results in a small group meeting, to more formal, stand-up affairs in front of rooms full of people. Regardless, if you have ever felt as though you were being tuned out or misunderstood by an audience, it may be time to learn to “see” through their eyes. In this short workshop, you’ll do a basic assessment of your preferred modes communication and identify some strategies for adjusting to audiences who have different cognitive styles and communication preferences from yours.

 

SC5: Using Statistical Engineering to Solve Large, Unstructured Problems

Thu, Feb 16, 1:30 PM - 5:30 PM
Fantail

Instructor(s): Ronald D Snee, Snee Associates, LLC

This course was designed to enhance the skills of statisticians in using statistical engineering to solve large, complex, unstructured problems encountered in business, industry, and government. We will identify important gaps in the theory of statistical engineering for which research is needed. Several case studies of the use of statistical engineering in a variety of fields will be presented. Issues to be addressed include understanding what statistical engineering is, why it is important, and how to use it, as well as identifying research gaps. We also will discuss how statistical engineering differs from the classic application of statistics. Participants will be introduced to the critical leadership skills needed for the successful use of statistical engineering. Each participant will develop a personal action plan for using statistical engineering in their work environment, whether in academia, government, or the private sector. Participants will gain insight into increasing the impact of their work and how to transition from being viewed as passive consultants to proactive leaders within their organizations. We will use presentation and discussion of material from articles about statistical engineering, as well as share personal experiences (participants and course leaders) in solving large, unstructured problems. The course will be highly interactive, enabling extensive participation by all.

 

SC6: Managing Your Time and Priorities

Thu, Feb 16, 1:30 PM - 5:30 PM
Zander

Instructor(s): Bill Williams, Organizational Learning Consultant

Time management isn’t a skill so much as a collection of around a dozen skills, from broader behaviors like goal setting to management skills like delegation, to more granular tactics like scheduling. Most of us excel at some and fall short on others. In this course, you will take a step back from the day-to-day to identify your most important priorities, look at what gets in the way of addressing them, identify your time management strengths and areas for improvement, and explore means of shoring up your weak areas.

 

Sat, Feb 18

T1: Putting Your Best Loafer Forward

Sat, Feb 18, 1:30 PM - 3:30 PM
Atlantis A

Instructor(s): Bill Williams, Organizational Learning Consultant

As professionals, the outcomes of our work are a critical part of our success. But we don't always give sufficient attention to how we achieve those outcomes and, particularly, the impressions co-workers, managers, clients, and others get from interacting with us. The purpose of this tutorial is to help you honestly assess what “foot” you put forward as a professional, enabling you to do the following: clarify the impressions you want others to have of you as a professional; and identify specific actions you can take to convey those impressions at work, in work-related activities after hours, and online.

 

T2: Promoting Your Consulting Career in the Era of Web 2.0

Sat, Feb 18, 1:30 PM - 3:30 PM
Bluegill

Instructor(s): Stephen David Simon, P.Mean Consulting

Web 2.0, defined by Wikipedia as "web applications that facilitate participatory information sharing, interoperability, user-centered design, and collaboration on the World Wide Web" offers new opportunities for you to promote your consulting career. These tools are mostly free or very inexpensive, though they are labor intensive. In this talk, I will describe some Web 2.0 tools that I have used to promote my independent consulting career (Facebook, Twitter, LinkedIn), as well as some of the unwritten rules about appropriate usage of these tools. I will contrast these tools with simpler Internet methods (static websites, email newsletters) and some non-computer methods that can help promote your consulting career. Web 2.0 will not replace more traditional modes of career promotion, but it offers some unique opportunities to supplement these efforts.

 

T3: Measurement Systems Analysis

Sat, Feb 18, 1:30 PM - 3:30 PM
Atlantis B

Instructor(s): Jennifer H. Van Mullekom, DuPont

Measurement Systems Analysis is critical to any development or improvement effort in your business. A poor measurement system can result in the inability to distinguish between product development candidates or the inability to determine the success of an improvement. This tutorial will provide basic instruction on how to complete and communicate measurement systems analysis for continuous variables, attribute (discrete) variables, and in-line systems. Multiple software packages will be demonstrated. Examples will include both transactional and engineering/R&D examples. The emphasis in this tutorial will include the practical aspects of designing, executing, analyzing, and communicating the measurement system study within the context of multi-disciplinary, cross-functional teams. Tutorial content prepared by Patrick DeFeo, DuPont.

 

T4: Bayesian Analysis in SAS

Sat, Feb 18, 1:30 PM - 3:30 PM
Crystal B

Instructor(s): Mike Patetta, SAS Institute

The tutorial focuses on Bayesian analyses using the PHREG, GENMOD, and MCMC procedures. Most of the examples are in the area of clinical trials. The specific topics that will be covered are: • Fit a logistic regression model in PROC GENMOD • Fit a survival model in PROC PHREG • Use prior distributions in a Bayesian analysis • Fit a logistic regression model, general linear mixed model, and zero-inflated Poisson model in PROC MCMC • Illustrate a Bayesian approach to clinical trials using PROC MCMC • Illustrate the Bayesian approach to meta-analysis

 
View Full Program