Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices

Detalhes bibliográficos
Autor(a) principal: Silva,Roberta Bessa Veloso
Data de Publicação: 2008
Outros Autores: Ferreira,Daniel Furtado, Nogueira,Denismar Alves
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Ciência e Agrotecnologia (Online)
Texto Completo: http://old.scielo.br/scielo.php?script=sci_arttext&pid=S1413-70542008000100023
Resumo: The present work emphasizes the importance of testing hypothesis on homogeneity of covariance matrices from multivariate k populations. The violation of the assumption of the homogeneity of covariance matrices affects the performance of the tests and the coverage probability of the confidence regions. This work intends to apply two tests of homogeneity of covariance and to evaluate type I error rates and power using Monte Carlo simulation in normal populations and robustness in non normal populations. Multivariate Bartlett's test (MBT) and its bootstrap version (MBTB) were used. Different configurations are tested combining sample sizes, number of variates, correlation and number of populations. Results show that the bootstrap test was considered superior to the asymptotic test and robust, since it controls the type error I rate.
id UFLA-2_7808722fb26a31094829c3b7bcac9c64
oai_identifier_str oai:scielo:S1413-70542008000100023
network_acronym_str UFLA-2
network_name_str Ciência e Agrotecnologia (Online)
repository_id_str
spelling Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matricesSimulationtype I error ratepower, Monte CarloThe present work emphasizes the importance of testing hypothesis on homogeneity of covariance matrices from multivariate k populations. The violation of the assumption of the homogeneity of covariance matrices affects the performance of the tests and the coverage probability of the confidence regions. This work intends to apply two tests of homogeneity of covariance and to evaluate type I error rates and power using Monte Carlo simulation in normal populations and robustness in non normal populations. Multivariate Bartlett's test (MBT) and its bootstrap version (MBTB) were used. Different configurations are tested combining sample sizes, number of variates, correlation and number of populations. Results show that the bootstrap test was considered superior to the asymptotic test and robust, since it controls the type error I rate.Editora da UFLA2008-02-01info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersiontext/htmlhttp://old.scielo.br/scielo.php?script=sci_arttext&pid=S1413-70542008000100023Ciência e Agrotecnologia v.32 n.1 2008reponame:Ciência e Agrotecnologia (Online)instname:Universidade Federal de Lavras (UFLA)instacron:UFLA10.1590/S1413-70542008000100023info:eu-repo/semantics/openAccessSilva,Roberta Bessa VelosoFerreira,Daniel FurtadoNogueira,Denismar Alveseng2008-03-12T00:00:00Zoai:scielo:S1413-70542008000100023Revistahttp://www.scielo.br/cagroPUBhttps://old.scielo.br/oai/scielo-oai.php||renpaiva@dbi.ufla.br|| editora@editora.ufla.br1981-18291413-7054opendoar:2022-11-22T16:30:16.726290Ciência e Agrotecnologia (Online) - Universidade Federal de Lavras (UFLA)true
dc.title.none.fl_str_mv Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
title Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
spellingShingle Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
Silva,Roberta Bessa Veloso
Simulation
type I error rate
power, Monte Carlo
title_short Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
title_full Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
title_fullStr Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
title_full_unstemmed Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
title_sort Robustness of asymptotic and bootstrap tests for multivariate homogeneity of covariance matrices
author Silva,Roberta Bessa Veloso
author_facet Silva,Roberta Bessa Veloso
Ferreira,Daniel Furtado
Nogueira,Denismar Alves
author_role author
author2 Ferreira,Daniel Furtado
Nogueira,Denismar Alves
author2_role author
author
dc.contributor.author.fl_str_mv Silva,Roberta Bessa Veloso
Ferreira,Daniel Furtado
Nogueira,Denismar Alves
dc.subject.por.fl_str_mv Simulation
type I error rate
power, Monte Carlo
topic Simulation
type I error rate
power, Monte Carlo
description The present work emphasizes the importance of testing hypothesis on homogeneity of covariance matrices from multivariate k populations. The violation of the assumption of the homogeneity of covariance matrices affects the performance of the tests and the coverage probability of the confidence regions. This work intends to apply two tests of homogeneity of covariance and to evaluate type I error rates and power using Monte Carlo simulation in normal populations and robustness in non normal populations. Multivariate Bartlett's test (MBT) and its bootstrap version (MBTB) were used. Different configurations are tested combining sample sizes, number of variates, correlation and number of populations. Results show that the bootstrap test was considered superior to the asymptotic test and robust, since it controls the type error I rate.
publishDate 2008
dc.date.none.fl_str_mv 2008-02-01
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://old.scielo.br/scielo.php?script=sci_arttext&pid=S1413-70542008000100023
url http://old.scielo.br/scielo.php?script=sci_arttext&pid=S1413-70542008000100023
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv 10.1590/S1413-70542008000100023
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv text/html
dc.publisher.none.fl_str_mv Editora da UFLA
publisher.none.fl_str_mv Editora da UFLA
dc.source.none.fl_str_mv Ciência e Agrotecnologia v.32 n.1 2008
reponame:Ciência e Agrotecnologia (Online)
instname:Universidade Federal de Lavras (UFLA)
instacron:UFLA
instname_str Universidade Federal de Lavras (UFLA)
instacron_str UFLA
institution UFLA
reponame_str Ciência e Agrotecnologia (Online)
collection Ciência e Agrotecnologia (Online)
repository.name.fl_str_mv Ciência e Agrotecnologia (Online) - Universidade Federal de Lavras (UFLA)
repository.mail.fl_str_mv ||renpaiva@dbi.ufla.br|| editora@editora.ufla.br
_version_ 1799874964515454976

Registros relacionados