On Stochastic Versions of the EM Algorithm - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 1995

On Stochastic Versions of the EM Algorithm

Gilles Celeux
Didier Chauveau

Résumé

We compare three different stochastic versions of the EM algorithm: The SEM algorithm, the SAEM algorithm and the MCEM algorithm. We suggest that the most relevant contribution of the MCEM methodology is what we call the simulated annealing MCEM algorithm, which turns out to be very close to SAEM. We focus particularly on the mixture of distributions problem. In this context, we review the available theoretical results on the convergence of these algorithms and on the behavior of SEM as the sample size tends to infinity. The second part is devoted to intensive Monte Carlo numerical simulations and a real data study. We show that, for some particular mixture situations, the SEM algorithm is almost always preferable to the EM and simulated annealing versions SAEM and MCEM. For some very intricate mixtures, however, none of these algorithms can be confidently used. Then, SEM can be used as an efficient data exploratory tool for locating significant maxima of the likelihood function. In the real data case, we show that the SEM stationary distribution provides a contrasted view of the loglikelihood by emphasizing sensible maxima.
Fichier principal
Vignette du fichier
RR-2514.pdf (409.78 Ko) Télécharger le fichier

Dates et versions

inria-00074164 , version 1 (24-05-2006)

Identifiants

  • HAL Id : inria-00074164 , version 1

Citer

Gilles Celeux, Didier Chauveau, Jean Diebolt. On Stochastic Versions of the EM Algorithm. [Research Report] RR-2514, INRIA. 1995. ⟨inria-00074164⟩
650 Consultations
4629 Téléchargements

Partager

Gmail Facebook X LinkedIn More