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

Authors:
José G. Dias;Michel Wedel
Affiliations:
Department of Quantitative Methods, Instituto Superior de Ciências do Trabalho e da Empresa—ISCTE, Av. das Forças Armadas, Lisboa 1649–026, Portugal. jose.dias@iscte. ...;The University of Michigan Business School, 701 Tappan Street, MI 48109 Ann Arbor, USA
Venue:
Statistics and Computing
Year:
2004

Citing 2
Cited 5

On convergence of the EM algorithmand the Gibbs sampler

Statistics and Computing
A Greedy EM Algorithm for Gaussian Mixture Learning

Neural Processing Letters

Adjusted Viterbi Training

Probability in the Engineering and Informational Sciences
Bayesian density estimation using skew student-t-normal mixtures

Computational Statistics & Data Analysis
Interval estimation in a finite mixture model: Modeling P-values in multiple testing applications

Computational Statistics & Data Analysis
Robust mixture modeling based on scale mixtures of skew-normal distributions

Computational Statistics & Data Analysis
Multivariate mixture modeling using skew-normal independent distributions

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.00

Visualization

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.