EVALUATION OF APPROXIMATED AND EXACT MULTIVARIATE TESTS FOR MEAN VECTORS: A DATA SIMULATION STUDY

The present study aimed to evaluate, through data simulation, the multivariate statistical tests Likelihood ratio test (LRT) and Hotelling’s T2 test for mean vectors regarding the type I error rate and the power of test. The scenarios were designed to analyze test performance under the influence of...

ver descrição completa

Detalhes bibliográficos
Autores: LACERDA, Maurício Silva, NASCIMENTO, Jhennifer dos Santos, BARBOSA, Eduardo Campana, MANULI, Rômulo César, NASCIMENTO, Moysés, NASCIMENTO, Ana Carolina Campana, EMILIANO, Paulo Cesar
Tipo de documento: artigo
Estado:Versão publicada
Data de publicação:2022
País:Brasil
Recursos:Universidade Federal de Lavras (UFLA)
Repositório:Brazilian Journal of Biometrics
Idioma:inglês
OAI Identifier:oai:biometria.ufla.br:article/560
Acesso em linha:https://biometria.ufla.br/index.php/BBJ/article/view/560
Access Level:Acceso aberto
Palavra-chave:Likelihood ratio test
Hotelling's T2
type I error
most powerful test
Descrição
Resumo:The present study aimed to evaluate, through data simulation, the multivariate statistical tests Likelihood ratio test (LRT) and Hotelling’s T2 test for mean vectors regarding the type I error rate and the power of test. The scenarios were designed to analyze test performance under the influence of p−variate normality, correlation, and homogeneity of variance, as well as number of variables and sample size. Our results show that the type I error rate was not affected by the violation of the assumptions of independence and homogeneity of variances, due to the presence of p−variate normality, differently from the power of test. In data simulation of p−variate distribution with heavier tails than usual (Student−t with 1 degree of freedom), the Hotelling’s T2 showed to be conservative, while the LRT showed better results, especially for small sample sizes.