Efficiency and convergence properties of slice samplers

被引：39

作者：

Mira, A

Tierney, L

机构：

[1] Univ Insubria, Fac Econ, I-21100 Varese, Italy

[2] Univ Minnesota, Minneapolis, MN 55455 USA

来源：

SCANDINAVIAN JOURNAL OF STATISTICS | 2002年 / 29卷 / 01期

关键词：

auxiliary variables; efficiency of MCMC; geometric ergodicity; Markov chain Monte Carlo; Metropolis-Hastings algorithm; Peskun ordering; uniform ergodicity;

D O I：

10.1111/1467-9469.00267

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：

The slice sampler (SS) is a method of constructing a reversible Markov chain with a specified invariant distribution. Given an independence Metropolis-Hastings algorithm (IMHA) it is always possible to construct a SS that dominates it in the Peskun sense. This means that the resulting SS produces estimates with a smaller asymptotic variance than the IMHA. Furthermore the SS has a smaller second-largest eigenvalue. This ensures faster convergence to the target distribution. A sufficient condition for uniform ergodicity of them SS is given and an: upper bound for the rate of convergence to stationarity is provided.

引用

页码：1 / 12

页数：12

共 19 条

[1] [Anonymous], 1998, THESIS U MINNESOTA
[2] BESAG J, 1993, J ROY STAT SOC B MET, V55, P25
[3] Damien P, 1999, J ROY STAT SOC B, V61, P331
[4] GENERALIZATION OF THE FORTUIN-KASTELEYN-SWENDSEN-WANG REPRESENTATION AND MONTE-CARLO ALGORITHM
EDWARDS, RG
SOKAL, AD
[J]. PHYSICAL REVIEW D, 1988, 38 (06): : 2009 - 2012
[5] Geyer CJ, 1992, STAT SCI, V7, P473, DOI [10.1214/ss/1177011137, DOI 10.1214/SS/1177011137]
[6] GEYER JC, 1998, UNPUB MARKOV CHAIN M
[7] GREEN PJ, 1997, J ROY STAT SOC B MET, V59, P511
[8] Higdon DM, 1998, J AM STAT ASSOC, V93, P585
[9] Metropolized independent sampling with comparisons to rejection sampling and importance sampling
Liu, JS
[J]. STATISTICS AND COMPUTING, 1996, 6 (02) : 113 - 119
[10] LIU JS, 1995, J ROY STAT SOC B MET, V57, P157

← 1 2 →