Computational Aspects of Fitting Mixture Models via the Expectation-Maximization Algorithm

Files in This Item:
File Description SizeFormat 
OHaganEtAl_Revised.pdf3.27 MBAdobe PDFDownload
Title: Computational Aspects of Fitting Mixture Models via the Expectation-Maximization Algorithm
Authors: O'Hagan, Adrian
Murphy, Thomas Brendan
Gormley, Isobel Claire
Permanent link:
Date: Dec-2012
Abstract: The Expectation–Maximization (EM) algorithm is a popular tool in a wide variety of statistical settings, in particular in the maximum likelihood estimation of parameters when clustering using mixture models. A serious pitfall is that in the case of a multimodal likelihood function the algorithm may become trapped at a local maximum, resulting in an inferior clustering solution. In addition, convergence to an optimal solution can be very slow. Methods are proposed to address these issues: optimizing starting values for the algorithm and targeting maximization steps efficiently. It is demonstrated that these approaches can produce superior outcomes to initialization via random starts or hierarchical clustering and that the rate of convergence to an optimal solution can be greatly improved.
Type of material: Journal Article
Publisher: Elsevier
Journal: Computational Statistics and Data Analysis
Volume: 56
Issue: 12
Start page: 3843
End page: 3864
Copyright (published version): 2012 Elsevier
Keywords: Convergence rateExpectation–maximization algorithmHierarchical clusteringModel-based clusteringMultimodal likelihood
DOI: 10.1016/j.csda.2012.05.011
Language: en
Status of Item: Peer reviewed 2015-09-25T11:55:37Z
Appears in Collections:Mathematics and Statistics Research Collection

Show full item record

Citations 20

Last Week
Last month
checked on Dec 12, 2018

Download(s) 50

checked on May 25, 2018

Google ScholarTM



This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.