Please use this identifier to cite or link to this item: http://hdl.handle.net/10216/112555
Author(s): Adelaide Figueiredo
Paulo Gomes
Title: Performance of the EM algorithm on the identification of a mixtureof Watson distributions defined on the hypersphere
Issue Date: 2006
Abstract: We consider a set of n individuals described by p standardised variables, and we suppose that the individuals are previouslyselected from a population and the variables are a sample of variablesassumed to come from a mixture of k bipolar Watson distributionsdefined on the hypersphere.In this context we provide the identification of the mixture through the EM algorithm and we also carry out a simulation study to comparethe maximum likelihood estimates obtained from samples of moderate size withthe respective asymptotic estimates. Our simulation results revealed goodperformance of the EM algorithm for moderate sample sizes.
Subject: Ciências Tecnológicas
Technological sciences
URI: http://hdl.handle.net/10216/112555
Document Type: Artigo em Revista Científica Internacional
Rights: restrictedAccess
Appears in Collections:FEP - Artigo em Revista Científica Internacional

Files in This Item:
File Description SizeFormat 
51037.1.docx16.25 kBMicrosoft Word XML    Request a copy
51037.pdf393.53 kBAdobe PDF    Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.