Efficient implementation of the Metropolis-Hastings algorithm, with application to the Cormack-Jolly-Seber model

被引：1

作者：

Link, William A. ^{[1
]}

Barker, Richard J. ^{[2
]}

机构：

[1] USGS Patuxent Wildlife Res Ctr, Laurel, MD 20708 USA

[2] Univ Otago, Dept Math & Stat, Dunedin, New Zealand

来源：

ENVIRONMENTAL AND ECOLOGICAL STATISTICS | 2008年 / 15卷 / 01期

关键词：

Cormack-Jolly-Seber model; Mark-recapture analysis; Markov chain Monte Carlo; Metropolis-Hastings algorithm;

D O I：

10.1007/s10651-007-0037-9

中图分类号：

X [环境科学、安全科学];

学科分类号：

08 ; 0830 ;

摘要：

Judicious choice of candidate generating distributions improves efficiency of the Metropolis-Hastings algorithm. In Bayesian applications, it is sometimes possible to identify an approximation to the target posterior distribution; this approximate posterior distribution is a good choice for candidate generation. These observations are applied to analysis of the Cormack-Jolly-Seber model and its extensions.

引用

页码：79 / 87

页数：9

共 6 条

[1] Bishop J.A., 1978, Philosophical Transactions of the Royal Society of London B Biological Sciences, V281, P517, DOI 10.1098/rstb.1978.0007
[2] BURNHAM KP, 1991, P TAIP S STAT I STAT
[3] ESTIMATES OF SURVIVAL FROM SIGHTING OF MARKED ANIMALS
CORMACK, RM
[J]. BIOMETRIKA, 1964, 51 (3-4) : 429 - &
[4] JOLLY GM, 1965, BIOMETRIKA, V52, P225, DOI 10.2307/2333826
[5] Modeling association among demographic parameters in analysis of open population capture-recapture data
Link, WA
Barker, RJ
[J]. BIOMETRICS, 2005, 61 (01) : 46 - 54
[6] SEBER GAF, 1965, BIOMETRIKA, V52, P249

← 1 →