Português
Ciência-IUL
Publications
Publication Detailed Description
An empirical comparison of EM, SEM and MCMC performance for problematic Gaussian mixture likelihoods
Journal Title
Statistics and Computing
Year (definitive publication)
2004
Language
English
Country
Netherlands
More Information
Web of Science®
Scopus
Google Scholar
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.
Acknowledgements
--
Keywords
Gaussian mixture models,EM algorithm,SEM algorithm,MCMC,Label switching,Loss functions,Conjugate prior,Hierarchical prior
Fields of Science and Technology Classification
- Mathematics - Natural Sciences
- Computer and Information Sciences - Natural Sciences
Funding Records
| Funding Reference | Funding Entity |
|---|---|
| SFRH/BD/890/2000 | Fundação para a Ciência e a Tecnologia |
