Model based labeling for mixture models

K-REx Repository

Show simple item record

dc.contributor.author Yao, Weixin
dc.date.accessioned 2012-05-17T16:04:36Z
dc.date.available 2012-05-17T16:04:36Z
dc.date.issued 2012-05-17
dc.identifier.uri http://hdl.handle.net/2097/13831
dc.description.abstract Label switching is one of the fundamental problems for Bayesian mixture model analysis. Due to the permutation invariance of the mixture posterior, we can consider that the posterior of a m-component mixture model is a mixture distribution with m! symmetric components and therefore the object of labeling is to recover one of the components. In order to do labeling, we propose to first fit a symmetric m!-component mixture model to the Markov chain Monte Carlo (MCMC) samples and then choose the label for each sample by maximizing the corresponding classification probabilities, which are the probabilities of all possible labels for each sample. Both parametric and semi-parametric ways are proposed to fit the symmetric mixture model for the posterior. Compared to the existing labeling methods, our proposed method aims to approximate the posterior directly and provides the labeling probabilities for all possible labels and thus has a model explanation and theoretical support. In addition, we introduce a situation in which the "ideally" labeled samples are available and thus can be used to compare different labeling methods. We demonstrate the success of our new method in dealing with the label switching problem using two examples. en_US
dc.relation.uri http://www.springerlink.com/content/k3477p14g6n53546/ en_US
dc.rights The final publication is available at www.springerlink.com en_US
dc.subject Bayesian mixtures en_US
dc.subject Labeling probabilities en_US
dc.subject Label switching en_US
dc.subject Markov chain Monte Carlo en_US
dc.subject Mixture model en_US
dc.title Model based labeling for mixture models en_US
dc.type Article (author version) en_US
dc.date.published 2012 en_US
dc.citation.doi doi:10.1007/s11222-010-9226-8 en_US
dc.citation.epage 347 en_US
dc.citation.issue 2 en_US
dc.citation.jtitle Statistics and Computing en_US
dc.citation.spage 337 en_US
dc.citation.volume 22 en_US
dc.contributor.authoreid wxyao en_US

Files in this item


Files Size Format View

This item appears in the following Collection(s)

Show simple item record