<b>Evaluating the power and the size of multivariate Levene´s test using Monte Carlo simulation and <em>bootstrap</em></b> - DOI: 10.4025/actascitechnol.v28i1.1293

Authors

  • Marcelo Angelo Cirillo UFLA
  • Daniel Furtado Ferreira UFLA
  • Thelma Sáfadi UFLA

DOI:

https://doi.org/10.4025/actascitechnol.v28i1.1293

Keywords:

simulação bootstrap, taxa de erro tipo I, poder do teste de Levene

Abstract

This work aimed to evaluate the I-type error rate and the power of Levene´s multivariate test using different location parameters to compare k covariance matrices. Monte Carlo´s simulation and bootstrap were used for this purpose. Configurations were considered with different degrees of correlations, heterogeneity of covariance matrices, sample sizes and number of variants. Results showed that Levene´s multivariate test centered on the mean is more powerfull in small sample size and with low degree of heterogeneity among covariance matrices; the bootstrap approach controlled the I-type error when median was used as scale parameter.

Downloads

Download data is not yet available.

Author Biography

Marcelo Angelo Cirillo, UFLA

Possui graduação em Bacharelado em Estatí­stica pela UNESP - Universidade Estadual Paulista Júlio de Mesquita Filho (1999), mestrado em Estatí­stica e Experimentação Agropecuária pela Universidade Federal de Lavras (2003) e doutorado em Estatí­stica e Experimentação Agropecuária pela Universidade Federal de Lavras (2006). Atualmente é professor Adjunto ní­vel I da Universidade Federal de Lavras atuando nas linhas de pesquisa: Análise Multivariada, Superfí­cie de Resposta e Estatí­stica Computacional com ênfase em métodos de computação intensiva Currí­culo Lattes

Downloads

Published

2008-03-20

How to Cite

Cirillo, M. A., Ferreira, D. F., & Sáfadi, T. (2008). <b>Evaluating the power and the size of multivariate Levene´s test using Monte Carlo simulation and <em>bootstrap</em></b> - DOI: 10.4025/actascitechnol.v28i1.1293. Acta Scientiarum. Technology, 28(1), 105–112. https://doi.org/10.4025/actascitechnol.v28i1.1293

Issue

Vol. 28 No. 1 (2006)

Section

Statistics

License

DECLARATION OF ORIGINALITY AND COPYRIGHTS

I Declare that current article is original and has not been submitted for publication, in part or in whole, to any other national or international journal. 

The copyrights belong exclusively to the authors. Published content is licensed under Creative Commons Attribution 4.0 (CC BY 4.0) guidelines, which allows sharing (copy and distribution of the material in any medium or format) and adaptation (remix, transform, and build upon the material) for any purpose, even commercially, under the terms of attribution.

Read this link for further information on how to use CC BY 4.0 properly.

 

Most read articles by the same author(s)