Método Bootstrap para hipótesis concernientes a la diferencia de medias para muestras pareadas: aplicaciones.

Abstract

In this paper , we propose a flexible method to calculate a confidence interval with bootstrap replicas for the mean difference of two paired samples without the need of normality, in the field of hypothesis theory. The null hypothesis is rejected if the confidence interval does not contain the 0, otherwise it is accepted . two applications are made with the variables : monthly milk production and average monthly milk production , from cows fed only with forage and with forage plus corn silage. a simulation study is also carried out with normal samples, obtaining similar results with the com- monly used method of t-student and the proposed bootstrap method.