Comparing Two Treatments in Two Period Repeated Measurement Design

Abstract

This work deals with nonparametric test procedures for comparing effects of two treatments in two period parallel group design, where each subject receives the same treatment over the two periods. The procedures are based on consistent estimators of relative treatment effects in each period and that of covariate adjusted effects with responses in period 1 as covariates. Related asymptotic results are obtained followed by necessary simulation studies to evaluate the performance of the tests. AMS 2010 subject classifications: 62G10 62J15