An adaptive version for the Metropolis Adjusted Langevin algorithm with a truncated drift

被引：80

作者：

Atchade, Yves F. ^{[1
]}

机构：

[1] Univ Ottawa, Dept Math & Stat, Ottawa, ON K1N 6N5, Canada

来源：

METHODOLOGY AND COMPUTING IN APPLIED PROBABILITY | 2006年 / 8卷 / 02期

基金：

加拿大自然科学与工程研究理事会;

关键词：

adaptive Markov Chain Monte Carlo; Langevin algorithms; Metropolis-Hastings algorithms; stochastic approximation algorithms;

D O I：

10.1007/s11009-006-8550-0

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

This paper extends some adaptive schemes that have been developed for the Random Walk Metropolis algorithm to more general versions of the Metropolis-Hastings (MH) algorithm, particularly to the Metropolis Adjusted Langevin algorithm of Roberts and Tweedie (1996). Our simulations show that the adaptation drastically improves the performance of such MH algorithms. We study the convergence of the algorithm. Our proves are based on a new approach to the analysis of stochastic approximation algorithms based on mixingales theory.

引用

页码：235 / 254

页数：20

共 20 条

[1] ANDRIEU C, 2005, IN PRESS ANN APPL PR
[2] ANDRIEU C, 2005, EFFICIENCY ADAPTIVE
[3] On adaptive Markov chain Monte Carlo algorithms
Atchadé, YF
Rosenthal, JS
[J]. BERNOULLI, 2005, 11 (05) : 815 - 828
[4] Renewal theory and computable convergence rates for geometrically ergodic Markov chains
Baxendale, PH
[J]. ANNALS OF APPLIED PROBABILITY, 2005, 15 (1B) : 700 - 738
[5] BENVENISTE A, 1990, APPL MATH
[6] Breyer L, 2002, TECHNICAL REPORT
[7] DAVIDSON J, 1997, ECONOMET REV, V16, P251
[8] Adaptive Markov chain Monte Carlo through regeneration
Gilks, WR
Roberts, GO
Suhu, SK
[J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1998, 93 (443) : 1045 - 1054
[9] An adaptive Metropolis algorithm
Haario, H
Saksman, E
Tamminen, J
[J]. BERNOULLI, 2001, 7 (02) : 223 - 242
[10] Hall P., 1980, Martingale Limit Theory and Its Application

← 1 2 →