The goal of this paper was to verify the influence of the dependency structure among variables on the power of multivariate combination-based permutation tests in multidimensional applications. On the basis of the results of the simulation study, correlation seems to affect combination-based permutation tests by reducing power of multivariate tests. Future prospects concern some specific procedures aimed at possibly improving power of multivariate combination-based permutation tests. The application of special forms of combination function known as truncated product method [7] is under investigation to verify if it is possible to mitigate the negative effect on the power of combination-based multivariate permutation tests produced by an increasing level of correlation/association among responses.
The Influence of the Dependency Structure in Combination-Based Multivariate Permutation Tests in Case of Ordered Categorical Responses
ARBORETTI GIANCRISTOFARO, ROSA;CARROZZO, ANNA ELEONORA;BOATTO, VASCO LADISLAO;BARISAN, LUIGINO
2014
Abstract
The goal of this paper was to verify the influence of the dependency structure among variables on the power of multivariate combination-based permutation tests in multidimensional applications. On the basis of the results of the simulation study, correlation seems to affect combination-based permutation tests by reducing power of multivariate tests. Future prospects concern some specific procedures aimed at possibly improving power of multivariate combination-based permutation tests. The application of special forms of combination function known as truncated product method [7] is under investigation to verify if it is possible to mitigate the negative effect on the power of combination-based multivariate permutation tests produced by an increasing level of correlation/association among responses.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.