Particle methods for maximum likelihood parameter estimation in latent variable models

AM Johansen, A Doucet, M Davy

Research output: Contribution to journalArticle (Academic Journal)peer-review

33 Citations (Scopus)

Abstract

Standard methods for maximum likelihood parameter estimation in latent variable models rely on the Expectation-Maximization algorithm and its Monte Carlo variants. Our approach is different and motivated by similar considerations to simulated annealing; that is we build a sequence of artificial distributions whose support concentrates itself on the set of maximum likelihood estimates. We sample from these distributions using a sequential Monte Carlo approach. We demonstrate state of the art performance for several applications of the proposed approach.
Translated title of the contributionParticle methods for maximum likelihood parameter estimation in latent variable models
Original languageEnglish
Pages (from-to)47 - 57
Number of pages11
JournalStatistics and Computing
Volume18 (1)
DOIs
Publication statusPublished - Mar 2008

Bibliographical note

Publisher: Springer

Fingerprint Dive into the research topics of 'Particle methods for maximum likelihood parameter estimation in latent variable models'. Together they form a unique fingerprint.

Cite this