Activity Number:
|
234
- Bayesian Conditional Models and Updates
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 31, 2017 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Bayesian Statistical Science
|
Abstract #323660
|
|
Title:
|
A Modified Conditional Metropolis-Hastings Sampler with Two Generalized Strategies
|
Author(s):
|
Jianan Hui* and James Flegal and Alicia Johnson
|
Companies:
|
University of California, Riverside and University of California, Riverside and Macalester College
|
Keywords:
|
Markov chain Monte Carlo ;
Metropolis-Hastings ;
Gibbs sampler ;
Geometric ergodicity
|
Abstract:
|
A modified conditional Metropolis-Hastings sampler for general state spaces is investigated under two generalized strategies. Under specified conditions, we show that the generalizations of the modified conditional Metropolis-Hastings sampler can also lead to substantial gains in statistical efficiency while maintaining the overall quality of convergence. Results are illustrated in both simulated and real data. For the simulated data settings, we use a toy bivariate Normal model and a Bayesian version of the random effects model. In order to illustrate its utility in high-dimensional simulations, we consider a dynamic space-time model on weather station data.
|
Authors who are presenting talks have a * after their name.