Confronto empirico tra la potenza dei test non parametrici per la verifica dell'uguaglianza delle leggi di distribuzione di due campioni indipendenti e i test di uniformità

Abstract

In this study a new procedure is proposed in order to transform the hypothesis of equality of distributional laws of two independent samples into an equivalent univariate goodness of fit hypothesis. The comparison of the power of two samples tests and the new uniformity tests here proposed is performed by simulations using Monte Carlo method.