An empirical comparison of EM, SEM and MCMC performance for problematic Gaussian mixture likelihoods

Show simple item record Dias, José G. en_US Wedel, Michel en_US 2006-09-11T19:36:33Z 2006-09-11T19:36:33Z 2004-10 en_US
dc.identifier.citation Dias, José G.; Wedel, Michel; (2004). "An empirical comparison of EM, SEM and MCMC performance for problematic Gaussian mixture likelihoods." Statistics and Computing 14(4): 323-332. <> en_US
dc.identifier.issn 0960-3174 en_US
dc.identifier.issn 1573-1375 en_US
dc.description.abstract We compare EM, SEM, and MCMC algorithms to estimate the parameters of the Gaussian mixture model. We focus on problems in estimation arising from the likelihood function having a sharp ridge or saddle points. We use both synthetic and empirical data with those features. The comparison includes Bayesian approaches with different prior specifications and various procedures to deal with label switching. Although the solutions provided by these stochastic algorithms are more often degenerate, we conclude that SEM and MCMC may display faster convergence and improve the ability to locate the global maximum of the likelihood function. en_US
dc.format.extent 530601 bytes
dc.format.extent 3115 bytes
dc.format.mimetype application/pdf
dc.format.mimetype text/plain
dc.language.iso en_US
dc.publisher Kluwer Academic Publishers; Springer Science+Business Media en_US
dc.subject.other Statistics en_US
dc.subject.other Data Structures, Cryptology and Information Theory en_US
dc.subject.other Numeric Computing en_US
dc.subject.other Mathematical Modeling and Industrial Mathematics en_US
dc.subject.other Statistics, General en_US
dc.subject.other Gaussian Mixture Models en_US
dc.subject.other EM Algorithm en_US
dc.subject.other SEM Algorithm en_US
dc.subject.other MCMC en_US
dc.subject.other Label Switching en_US
dc.subject.other Loss Functions en_US
dc.subject.other Conjugate Prior en_US
dc.subject.other Hierarchical Prior en_US
dc.title An empirical comparison of EM, SEM and MCMC performance for problematic Gaussian mixture likelihoods en_US
dc.type Article en_US
dc.subject.hlbsecondlevel Mathematics en_US
dc.subject.hlbsecondlevel Statistics and Numeric Data en_US
dc.subject.hlbtoplevel Social Sciences en_US
dc.subject.hlbtoplevel Science en_US
dc.description.peerreviewed Peer Reviewed en_US
dc.contributor.affiliationum The University of Michigan Business School, 701 Tappan Street, MI, 48109, Ann Arbor, USA en_US
dc.contributor.affiliationother Department of Quantitative Methods, Instituto Superior de Ciências do Trabalho e da Empresa—ISCTE, Av. das Forças Armadas, Lisboa, 1649–026, Portugal en_US
dc.contributor.affiliationumcampus Ann Arbor en_US
dc.description.bitstreamurl en_US
dc.identifier.doi en_US
dc.identifier.source Statistics and Computing en_US
dc.owningcollname Interdisciplinary and Peer-Reviewed
 Show simple item record

This item appears in the following Collection(s)

Search Deep Blue

Advanced Search

Browse by

My Account


Available Now

MLibrary logo