A Nonparametric Approach to the Analysis of the Two-Treatment, Two-Period, Four-Sequence Crossover Model

Publisher Website
Google Scholar
Abstract
Owing to the estimability problesm in the two-treatment, two-period, two-sequence crossover model, the four-sequence design (AB,BA,AA,BB) has gained popularity. This paper presents a nonparametric approach to the analysis of the two-treatment, two-period, four-sequence crossover model, using an approach similar to that of Koch (1972, Biometrics 28, 577-584). After forming differences within an experimental unit, the Benard-van Elteren test (Bernard and van Elteren, 1953, Proc. Komink. Nederlandse Akad van Wetenshappen, Series A 56, 358-369 uses the transformed data to construct a test of the equality of the direct effects of the treatment. A numerical example is presented to illustrate the methodology.

This publication has 1 reference indexed in Scilit: