The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
47
|
Type:
|
Invited
|
Date/Time:
|
Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
|
Sponsor:
|
Section on Bayesian Statistical Science
|
Abstract - #303636 |
Title:
|
Self-Learning Markov Chain Monte Carlo from a Large Deviations Point of View
|
Author(s):
|
Jingchen Liu*+ and Xuan Yang and Henrik Hult
|
Companies:
|
Columbia University and Columbia University and Royal Institute of Technology, Sweden
|
Address:
|
1255 Amsterdam Avenue, New York, NY, , United States
|
Keywords:
|
MCMC ;
large deviations ;
Adaptive Monte Carlo
|
Abstract:
|
Measuring the convergence of Markov chains and comparing the efficiency among different MCMC schemes have been a long lasting problem. In this talk, we propose using the large deviations rate function as an efficiency measure of an MCMC scheme. This measure turns out to be consistent with the existing asymptotic results in literature. In addition, the proposed efficiency measure can be evaluated numerically by the Markov chain itself and thus can be combined with adaptive MCMC schemes.
|
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 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.