Pre & Post Conference Short Courses

Two short courses will be offered on Wednesday, October 11th, and two on Saturday, October 14th from 8:30 a.m. to 5:30 p.m. The fee for each course includes coffee breaks and lunch. REGISTRATION IS LIMITED!!

Hands-On Bayesian Data Analysis Using WinBUGS by William F. Guthrie ($250) – Wednesday, October 11th
This workshop is designed to provide statisticians, scientists, or engineers with the tools necessary to begin to easily use Bayesian inference in applied problems. Participants in the course will learn the basics of Bayesian modeling and inference using Markov chain Monte Carlo simulation with the open-source software package Win BUGS. The workshop will introduce some of the theory underlying Bayesian analysis, but will primarily focus on Bayesian analysis of “real-world” scientific applications using examples from collaborative research with NIST scientists and engineers. Topics discussed will include Bayesian modeling, Markov chain Monte Carlo algorithms, convergence tests, model validation and inference. No knowledge of Bayesian methods is assumed, but experience analyzing different types of data will be helpful. The workshop will be very hands-on. Laptop computers will be available for each participant (or bring your own), and will be installed with data and software to be used throughout the day. Each participant will also receive a CD with all examples and exercises.

Statistical Consulting and Change Management Workshop by Sue Ellen Bisgaard and Soren Bisgaard ($250) – Wednesday, October 11th.
How do I identify the real problem? Why would a client lie to me? What if I don’t know the answer? Why did they hire me if they don’t want to do what I tell them to do? Just because it does not appear rational to you does not mean that the client irrational. What does a consulting contract look like? How do I handle confidential information? How do I charge for my services? Effective consulting involves more than having technical expert knowledge. Every system has a unique culture that may appear to be trying to block you. The goal of this workshop is to help you identify underlying manifestations of resistance and understand some approaches that may help you overcome, intervene and be more effective as a statistical consultant. We will also provide you with a number of practical tips based on our experience gained from consulting in many different industries and cultures.

Logistic and Poisson Regression with Applications by Douglas C. Montgomery ($250) – Saturday, October 14th.
Many modeling building and data analysis problems involve responses that are either binary or counts. Examples include customer satisfaction studies, failure analysis, risk modeling and financial decision making, and dose-response studies in the pharmaceutical industry. The analysis of this type of data often involves logistic regression for the binary case and Poisson regression for counts. This course is an introduction to these techniques. Participants should have a basic working knowledge of linear regression. Minitab and JMP will be featured prominently, and course participants will develop a good understanding of how these techniques are implemented in these two software packages. Applications are drawn from a variety of industrial and business settings. Course topics will include a review of linear regression fundamentals, some key topics from nonlinear models, binary responses – the logistic regression model, fitting logistic regression models, Wald inference and likelihood inference, model diagnostics, nominal and ordinal logistic regression, Poisson regression model fitting, checking, and inference; and, overdispersion.

*CANCELLED* Concrete R for Statistics and Quality Improvement by Antonio Possolo ($250) – Saturday, October 14th.
This course provides an introduction to statistical computation using R for quality professionals and statisticians. R is free software, both in the sense that no fees or registration are required to download it from http://www.r-project.org or to use it. This course will be particularly useful to those who contemplate migrating from a point-and-click interface to an accessible, flexible programming environment for statistical computation and graphics. A series of complete, self-contained examples of R at work to solve concrete statistical problems, involving applications of statistical methods in a wide range of areas in science and technology will be presented. Participants should be familiar with standard statistical methods and models, such as graphical and exploratory data analysis, analysis of variance, linear and non-linear regression, time series analysis, and experimental design. Participants will have ample opportunity to use R themselves throughout the course. Laptop computers, and all necessary supporting materials (software, documentation, and data) will be made available for individual use. An Introduction to R, by W.N. Venables, D.M. Smith, and the R Development Core Team (freely available at http://cran.r-project.org/doc/manuals/R-intro.pdf) provides a leisurely, useful overview of R.