Title:  Intermediate Bayesian Data Analysis Using WinBUGS and BRugs

Instructor:  Brad Carlin, University of Minnesota

Abstract:  Most statisticians and health policy researchers have been exposed
to Bayesian methods, and have some idea about the hierarchical modeling and
other settings in which they offer substantial benefits.  But actually
obtaining these benefits remains out of reach for many, due to a lack
of experience with modern Bayesian software in the analysis of real
data.  In this workshop, we will offer a hands-on opportunity to
explore the use of WinBUGS, the leading Bayesian software package, in
a variety of important models, with special attention paid to clinical
trial design and analysis, and to joint models for longitudinal
measurements and survival data.   Basic elements such as model
building, MCMC convergence diagnosis and acceleration, and posterior
plotting and summarization will be mentioned, as well as (time
permitting) important data-analytic procedures such as residual
analysis, model adequacy (through Bayesian p-values and CPO
statistics), variable selection, and model choice (through posterior
probabilities and DIC statistics).  In addition to WinBUGS, we will
also provide a brief introduction to BRugs, the new version of BUGS
available directly within the popular R package, which enables
simultaneous use of the features of both languages.  

Students should bring their own laptop computers to the session, and
to have the latest versions of WinBUGS and R already installed on
these computers. Both of these programs are freely available from 

http://www.mrc-bsu.cam.ac.uk/bugs/winbugs/contents.shtml

and

http://www.r-project.org/

respectively.  The presentation will assume familiarity with basic
Bayesian methods and MCMC algorithms, at the level of, say, Chapters 2
and 5 of Carlin and Louis (2000) or Chapters 2, 3, 5 and 11 of Gelman
et al. (2004).  The workshop's goal is to make these methods come
alive in the software through real data examples that the students try
for themselves during the presentation.  All necessary WinBUGS and
BRugs code will be made available on the web, with the joint modeling
material located at 

http://www.biostat.umn.edu/~brad/software.html