|
Activity Number:
|
247
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Council of Chapters
|
| Abstract - #300219 |
|
Title:
|
Center-Adjusted Inference for a Nonparametric Bayesian Random Effect Distribution
|
|
Author(s):
|
Yisheng Li*+ and Peter Mueller and Xihong Lin
|
|
Companies:
|
The University of Texas M.D. Anderson Cancer Center and The University of Texas M.D. Anderson Cancer Center and Harvard University
|
|
Address:
|
1515 Holcombe Blvd., Unit 447, Houston, TX, 77030,
|
|
Keywords:
|
Fixed effects ; Generalized linear mixed model ; Moments of a Dirichlet process ; Post processing ; Random effects ; Random probability measure
|
|
Abstract:
|
Dirichlet process (DP) prior models are a popular choice for semiparametric Bayesian general or generalized linear mixed models. The base measure of the DP prior for the random effect distribution is often specified as a normal distribution with a zero mean. The fact that this prior allows the random effect distribution to have a nonzero mean complicates the interpretation of inference for the fixed effects paired with the random effects. The resulting inference for the fixed effects may be poor. We propose a post-processing technique to adjust the inference. The approach uses a parametrization of the DP with a base measure centered at an unknown nonzero mean. The proposed adjustment for the analytically evaluated moments of the DP can be conveniently incorporated into the MCMC simulations. We evaluate the performance of the method through simulations and apply it to a real data set.
|
