|
Activity Number:
|
531
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:30 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #305110 |
|
Title:
|
Sequential Monte Carlo Samplers for Bayesian Generalized Linear Mixed Models
|
|
Author(s):
|
Yi-Ju Chen*+ and Yi-Liang Tung
|
|
Companies:
|
University of Wisconsin-Madison and Consultant
|
|
Address:
|
603 Eagle Heights Apt. C, Madison, WI, 53705,
|
|
Keywords:
|
Bayesian inference ; Generalized linear mixed model ; Particle filter algorithm ; Rao-Blackwellization ; Sequential imputation ; Sequential Monte Carlo
|
|
Abstract:
|
Sequential Monte Carlo samplers are developed for fitting Bayesian generalized linear mixed models with a particular prior specification. Explicitly, we propose sequential imputation methods for the involved posterior computations. The proposed samplers can not only simulate identically distributed draws from the posterior of interest, but also can be viewed as a Rao-Blackwellized version of particle filter algorithms. Finally, we illustrate the versatility of the samplers in dealing with different types of response variables with three empirical examples.
|
