A SLOWLY MIXING MARKOV-CHAIN WITH IMPLICATIONS FOR GIBBS SAMPLING

被引：16

作者：

MATTHEWS, P ^{[1
]}

机构：

[1] UNIV MARYLAND,BALTIMORE,MD 21201

来源：

STATISTICS & PROBABILITY LETTERS | 1993年 / 17卷 / 03期

关键词：

GIBBS SAMPLING; POSTERIOR DISTRIBUTION; MIXING RATE; COUPLING;

D O I：

10.1016/0167-7152(93)90172-F

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：

We give a Markov chain that converges to its stationary distribution very slowly. It has the form of a Gibbs sampler running on a posterior distribution of a parameter theta given data X. Consequences for Gibbs sampling are discussed.

引用

页码：231 / 236

页数：6

共 10 条

[1]

APPLEGATE D, 1991, SAMPLING INTEGRATION

[2]

CUI LM, 1992, MONITORING CONVERGEN

[3]

Doob J. L., 1953, STOCHASTIC PROCESSES

[4]

DOSS H, 1991, STUDY CONVERGENCE PR

[5]

DYER M, 1991, 91104 CARN MELL U DE

[6] SAMPLING-BASED APPROACHES TO CALCULATING MARGINAL DENSITIES [J].

GELFAND, AE ;

SMITH, AFM .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1990, 85 (410) :398-409

[7] ILLUSTRATION OF BAYESIAN-INFERENCE IN NORMAL DATA MODELS USING GIBBS SAMPLING [J].

GELFAND, AE ;

HILLS, SE ;

RACINEPOON, A ;

SMITH, AFM .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1990, 85 (412) :972-985

[8]

GELMAN A, 1992, HONEST INFERENCES IT

[9] LARGE CLIQUES ELUDE THE METROPOLIS PROCESS [J].

JERRUM, M .

RANDOM STRUCTURES & ALGORITHMS, 1992, 3 (04) :347-359

[10]

[No title captured]

← 1 →