Bayesian inference for Gibbs random fields using composite likelihoods
N. Friel. Proceedings of the Winter Simulation Conference, page 28:1--28:8. Winter Simulation Conference, (2012)
Abstract
Gibbs random fields play an important role in statistics, for example the autologistic model is commonly used to model the spatial distribution of binary variables defined on a lattice. However they are complicated to work with due to an intractability of the likelihood function. It is therefore natural to consider tractable approximations to the likelihood function. Composite likelihoods offer a principled approach to constructing such approximation. The contribution of this paper is to examine the performance of a collection of composite likelihood approximations in the context of Bayesian inference.
Description
Bayesian inference for gibbs random fields using composite likelihoods
%0 Conference Paper
%1 friel2012bayesian
%A Friel, Nial
%B Proceedings of the Winter Simulation Conference
%D 2012
%I Winter Simulation Conference
%K Bayesian Markov_random_field composite_likelihood statistics
%P 28:1--28:8
%T Bayesian inference for Gibbs random fields using composite likelihoods
%U http://dl.acm.org/citation.cfm?id=2429759.2429795
%X Gibbs random fields play an important role in statistics, for example the autologistic model is commonly used to model the spatial distribution of binary variables defined on a lattice. However they are complicated to work with due to an intractability of the likelihood function. It is therefore natural to consider tractable approximations to the likelihood function. Composite likelihoods offer a principled approach to constructing such approximation. The contribution of this paper is to examine the performance of a collection of composite likelihood approximations in the context of Bayesian inference.
@inproceedings{friel2012bayesian,
abstract = {Gibbs random fields play an important role in statistics, for example the autologistic model is commonly used to model the spatial distribution of binary variables defined on a lattice. However they are complicated to work with due to an intractability of the likelihood function. It is therefore natural to consider tractable approximations to the likelihood function. Composite likelihoods offer a principled approach to constructing such approximation. The contribution of this paper is to examine the performance of a collection of composite likelihood approximations in the context of Bayesian inference.},
acmid = {2429795},
added-at = {2013-05-15T16:21:24.000+0200},
articleno = {28},
author = {Friel, Nial},
biburl = {https://www.bibsonomy.org/bibtex/2db046b1a94282e5a84ebc41c19f27af3/peter.ralph},
booktitle = {Proceedings of the Winter Simulation Conference},
description = {Bayesian inference for gibbs random fields using composite likelihoods},
interhash = {537ac87535fd00e07a792486a155c678},
intrahash = {db046b1a94282e5a84ebc41c19f27af3},
keywords = {Bayesian Markov_random_field composite_likelihood statistics},
location = {Berlin, Germany},
numpages = {8},
pages = {28:1--28:8},
publisher = {Winter Simulation Conference},
series = {WSC '12},
timestamp = {2013-05-15T16:21:24.000+0200},
title = {Bayesian inference for {Gibbs} random fields using composite likelihoods},
url = {http://dl.acm.org/citation.cfm?id=2429759.2429795},
year = 2012
}