|
Activity Number:
|
489
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 7, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #300098 |
|
Title:
|
The Sample Metropolis-Hastings Algorithm
|
|
Author(s):
|
Chuanhai Liu+ and Andrew Lewandowski*+
|
|
Companies:
|
Purdue University and Purdue University
|
|
Address:
|
Department of Statistics, West Lafayette, IN, 47907-2066, Department of Statistics, West Lafayette, IN, 47907-2066,
|
|
Keywords:
|
Bayesian argument ; Deconvolution ; Dempster-Shafer Theory ; Population Monte Carlo ; Stochastic Approximation
|
|
Abstract:
|
The Metropolis-Hastings (MH) algorithm is a powerful tool used to derive most Markov chain Monte Carlo (MCMC) sampling schemes for Bayesian computation. In the past decade, researchers have introduced modifications to the MH algorithm such as Population Monte Carlo (PMC) and Stochastic Approximation Monte Carlo (SAMC) in an attempt to effectively use information from past samples. In the tradition of these methods, the Sample Metropolis-Hastings (SMH) algorithm is a MH-based algorithm which creates updates based on a stored sample of values. Examples and theoretical properties are discussed, and the SMH algorithm is compared to similar methods, such as MH, PMC, and SAMC.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2008 program |
