Show simple item record

dc.contributor.authorCasella, George
dc.contributor.authorRobert, Christian P.
dc.contributor.authorWells, Martin T.
dc.date.accessioned2011-04-29T13:58:24Z
dc.date.available2011-04-29T13:58:24Z
dc.date.issued2004
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/6114
dc.language.isoenen
dc.subjectMonte Carlo methodsen
dc.subjectBayes estimationen
dc.subjectPartition decompositionen
dc.subjectPosterior probabilitiesen
dc.subjectGibbs samplingen
dc.subject.ddc519en
dc.subject.classificationjelC11en
dc.subject.classificationjelC15en
dc.titleMixture models, latent variables and partitioned importance samplingen
dc.typeArticle accepté pour publication ou publié
dc.description.abstractenGibbs sampling has had great success in the analysis of mixture models. In particular, the “latent variable” formulation of the mixture model greatly reduces computational complexity. However, one failing of this approach is the possible existence of almost-absorbing states, called trapping states, as it may require an enormous number of iterations to escape from these states. Here we examine an alternative approach to estimation in mixture models, one based on a Rao–Blackwellization argument applied to a latent-variable-based estimator. From this derivation we construct an alternative Monte Carlo sampling scheme that avoids trapping states.en
dc.relation.isversionofjnlnameStatistical Methodology
dc.relation.isversionofjnlvol1en
dc.relation.isversionofjnlissue1-2en
dc.relation.isversionofjnldate2004
dc.relation.isversionofjnlpages1-18en
dc.relation.isversionofdoihttp://dx.doi.org/10.1016/j.stamet.2004.05.001en
dc.description.sponsorshipprivateouien
dc.relation.isversionofjnlpublisherElsevieren
dc.subject.ddclabelProbabilités et mathématiques appliquéesen


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record