Show simple item record

Weakly informative reparameterisations for location-scale mixtures

hal.structure.identifierCEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
dc.contributor.authorKamary, Kaniav
HAL ID: 179274		*
hal.structure.identifier
dc.contributor.authorLee, Jeong Eun*
hal.structure.identifierCEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
dc.contributor.authorRobert, Christian P.*
dc.date.accessioned2017-10-30T14:04:49Z
dc.date.available2017-10-30T14:04:49Z
dc.date.issued2018
dc.identifier.issn1061-8600
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/16868
dc.language.isoenen
dc.subjectNon-informative prior
dc.subjectimproper prior
dc.subjectmixture of distributions
dc.subjectBayesian analysis
dc.subjectDirichlet prior
dc.subjectexchangeability
dc.subjectpolar coordinates
dc.subjectcompound distributions
dc.subject.ddc519en
dc.titleWeakly informative reparameterisations for location-scale mixtures
dc.typeArticle accepté pour publication ou publié
dc.description.abstractenWhile mixtures of Gaussian distributions have been studied for more than a century (Pearson, 1894), the construction of a reference Bayesian analysis of those models still remains unsolved, with a general prohibition of the usage of improper priors (Fruwirth-Schnatter, 2006) due to the ill-posed nature of such statistical objects. This difficulty is usually bypassed by an empirical Bayes resolution (Richardson and Green, 1997). By creating a new parameterisation cantered on the mean and possibly the variance of the mixture distribution itself, we are able to develop here a weakly informative prior for a wide class of mixtures with an arbitrary number of components. We demonstrate that some posterior distributions associated with these priors is almost surely proper and we provide MCMC implementations that exhibit the expected exchangeability. We only study here the univariate case, the extension to multivariate location-scale mixtures being currently under study. An R package called Ultimixt is attached to this paper.
dc.relation.isversionofjnlnameJournal of Computational and Graphical Statistics
dc.relation.isversionofjnlvol27
dc.relation.isversionofjnlissue4
dc.relation.isversionofjnldate2018
dc.relation.isversionofjnlpages836-848
dc.relation.isversionofdoi10.1080/10618600.2018.1438900
dc.identifier.urlsitehttps://arxiv.org/abs/1601.01178
dc.relation.isversionofjnlpublisherTaylor & Francis
dc.subject.ddclabelProbabilités et mathématiques appliquéesen
dc.description.ssrncandidatenon
dc.description.halcandidatenon
dc.description.readershiprecherche
dc.description.audienceInternational
dc.relation.Isversionofjnlpeerreviewedoui
dc.date.updated2019-03-25T15:12:23Z
hal.author.functionaut
hal.author.functionaut
hal.author.functionaut


Files in this item

Thumbnail
Name:
1601.01178.pdf
Size:
3.637Mb
Format:
PDF
View/Open

This item appears in the following Collection(s)

Show simple item record