MCMC algorithms for constrained variance matrices

被引:26
作者
Browne, WJ [1 ]
机构
[1] Univ Nottingham, Sch Math Sci, Nottingham NG7 2RD, England
关键词
adaptive Metropolis-Hastings sampler; multi-level modelling; multi-vatiate responses; hierarchical modelling; Markov chain Monte Carlo (MCMC); MCMC efficiency;
D O I
10.1016/j.csda.2005.02.008
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
The problem of finding a generic algorithm for applying Markov chain Monte Carlo (MCMC) estimation procedures to statistical models that include variance matrices with additional parameter constraints is considered. Such problems can be split between additional constraints across variance matrices and within variance matrices. The case of additional constraints across variance matrices is considered here for the first time and a review of existing work on the case of additional parameter constraints within a variance matrix is given. Two simple single-site updating random walk Metropolis algorithms are described which have the advantage of generality in that they can be applied to virtually all scenarios. Four applications where these methods can be used in practice are given. Some situations when such single-site algorithms break down are described and multiple-site alternatives are briefly discussed. (C) 2005 Elsevier B.V. All rights reserved.
引用
收藏
页码:1655 / 1677
页数:23
相关论文
共 28 条
[1]   BAYESIAN-ANALYSIS OF BINARY AND POLYCHOTOMOUS RESPONSE DATA [J].
ALBERT, JH ;
CHIB, S .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1993, 88 (422) :669-679
[2]  
[Anonymous], 2003, MCMC ESTIMATION MLWI
[3]   Implementation and performance issues in the Bayesian and likelihood fitting of multilevel models [J].
Browne, WJ ;
Draper, D .
COMPUTATIONAL STATISTICS, 2000, 15 (03) :391-420
[4]   Bayesian and likelihood methods for fitting multilevel models with complex level-1 variation [J].
Browne, WJ ;
Draper, D ;
Goldstein, H ;
Rasbash, J .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2002, 39 (02) :203-225
[5]  
BROWNE WJ, THESIS U BATH, P98
[6]   BAYES INFERENCE IN THE TOBIT CENSORED REGRESSION-MODEL [J].
CHIB, S .
JOURNAL OF ECONOMETRICS, 1992, 51 (1-2) :79-99
[7]   Bayesian analysis of cross-section and clustered data treatment models [J].
Chib, S ;
Hamilton, BH .
JOURNAL OF ECONOMETRICS, 2000, 97 (01) :25-50
[8]   Analysis of multivariate probit models [J].
Chib, S ;
Greenberg, E .
BIOMETRIKA, 1998, 85 (02) :347-361
[9]   UNDERSTANDING THE METROPOLIS-HASTINGS ALGORITHM [J].
CHIB, S ;
GREENBERG, E .
AMERICAN STATISTICIAN, 1995, 49 (04) :327-335
[10]  
DRAPER D, 1997, UNPUB PRACTICAL MCMC