A note on EM algorithm for mixture models

dc.citation.doi	doi:10.1016/j.spl.2012.10.017	en_US
dc.citation.epage	526	en_US
dc.citation.issue	2	en_US
dc.citation.jtitle	Statistics and Probability Letters	en_US
dc.citation.spage	519	en_US
dc.citation.volume	83	en_US
dc.contributor.author	Yao, Weixin
dc.contributor.authoreid	wxyao	en_US
dc.date.accessioned	2013-01-22T17:24:19Z
dc.date.available	2013-01-22T17:24:19Z
dc.date.issued	2013-01-22
dc.date.published	2013	en_US
dc.description.abstract	Expectation-maximization (EM) algorithm has been used to maximize the likelihood function or posterior when the model contains unobserved latent variables. One main important application of EM algorithm is to find the maximum likelihood estimator for mixture models. In this article, we propose an EM type algorithm to maximize a class of mixture type objective functions. In addition, we prove the monotone ascending property of the proposed algorithm and discuss some of its applications.	en_US
dc.identifier.uri	http://hdl.handle.net/2097/15224
dc.language.iso	en_US	en_US
dc.relation.uri	http://www.sciencedirect.com/science/article/pii/S0167715212003896	en_US
dc.subject	Adaptive regression	en_US
dc.subject	EM algorithm	en_US
dc.subject	Edge-preserving smoothers	en_US
dc.subject	Mode	en_US
dc.subject	Robust regression	en_US
dc.title	A note on EM algorithm for mixture models	en_US
dc.type	Article (author version)	en_US

Files

Now showing 1 - 1 of 1

Now showing 1 - 1 of 1