Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
Implementations of the Monte Carlo EM algorithm
Autore:
Levine, RA; Casella, G;
Indirizzi:
Univ Calif Davis, Dept Stat, Davis, CA 95616 USA Univ Calif Davis Davis CA USA 95616 Davis, Dept Stat, Davis, CA 95616 USA Univ Florida, Dept Stat, Gainesville, FL 32611 USA Univ Florida Gainesville FL USA 32611 ept Stat, Gainesville, FL 32611 USA Cornell Univ, Dept Biomet, Ithaca, NY 14853 USA Cornell Univ Ithaca NY USA 14853 Univ, Dept Biomet, Ithaca, NY 14853 USA Cornell Univ, Dept Stat Sci, Ithaca, NY 14853 USA Cornell Univ Ithaca NY USA 14853 niv, Dept Stat Sci, Ithaca, NY 14853 USA
Titolo Testata:
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
fascicolo: 3, volume: 10, anno: 2001,
pagine: 422 - 439
SICI:
1061-8600(200109)10:3<422:IOTMCE>2.0.ZU;2-W
Fonte:
ISI
Lingua:
ENG
Soggetto:
EXPLORING POSTERIOR DISTRIBUTIONS; LINEAR MIXED MODELS; MARKOV-CHAINS; CONVERGENCE; BINARY;
Keywords:
generalized linear mixed models; Gibbs sampler; importance sampling; Markov chain Monte Carlo; Metropolis-Hastings algorithm; regenerative simulation; renewal theory;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Physical, Chemical & Earth Sciences
Citazioni:
25
Recensione:
Indirizzi per estratti:
Indirizzo: Levine, RA Univ Calif Davis, Dept Stat, 1 Shields Ave, Davis, CA 95616 USAUniv Calif Davis 1 Shields Ave Davis CA USA 95616 CA 95616 USA
Citazione:
R.A. Levine e G. Casella, "Implementations of the Monte Carlo EM algorithm", J COMPU G S, 10(3), 2001, pp. 422-439

Abstract

The Monte Carlo EM (MCEM) algorithm is a modification of the EM algorithm where the expectation in the E-step is computed numerically through Monte Carlo simulations. The most flexible and generally applicable approach to obtaining a Monte Carlo sample in each iteration of an MCEM algorithm is through Markov chain Monte Carlo (MCMC) routines such as the Gibbs and Metropolis-Hastings samplers. Although MCMC estimation presents a tractable solution to problems where the E-step is not available in closed form, two issues arise when implementing this MCEM routine: (1) how do we minimize the computational cost in obtaining an MCMC sample? and (2) how do we choose the Monte Carlo sample size? We address the first question through an application of importance sampling whereby samples drawn during previous EM iterations are recycled rather than running an MCMC sampler each MCEM iteration. The second question is addressed through an application of regenerative simulation. We obtain approximate independent and identical samples by subsampling the generated MCMC sample during different renewal periods. Standard central limit theorems may thus be used to gauge Monte Carlo error. In particular, we apply an automated rule for increasing the Monte Carlo sample size when the Monte Carlo error overwhelms the EM estimate at any given iteration. We illustrate our MCEM algorithm through analyses of two datasets fit by generalized linear mixed models. As a part of these applications, we demonstrate the improvement in computational cost and efficiency of our routine over alternative MCEM strategies.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 03/04/20 alle ore 07:21:26