NASA NTRS · 19850007946
Bayesian estimation of normal mixture parameters
Abstract
A Bayesian, or penalized maximum likelihood, approach to the problem of estimating the parameters of a mixture of multivariate normal distributions is proposed. The Bayesian formulation eliminates the problem of singularities in the likelihood function and results in an attractive EM-like procedure. Although the question of consistency is not settled, it is suggested that the proposed method has certain advantages over both the constrained and unconstrained maximum likelihood procedures.
Keep this discovery
Explore connections, maps & timelines
Peters, C.. 1984-01-01. Bayesian estimation of normal mixture parameters. https://searchnasa.com/record/nasa%3A19850007946/bayesian-estimation-of-normal-mixture-parameters
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Search NASA by Indiciae: BibTeX · RIS.