Nonparametric approaches for comparing three-period, two-treatment, four-sequence crossover designs. Issue 7 (3rd May 2019)

• Record Type: Journal Article
• Authors: Chatterjee, Suryasish; Bandyopadhyay, Uttam
• Is Part Of: Journal of statistical computation and simulation. Volume 89:Issue 7(2019)
• Page Start: 1153
• Page End: 1182
• Publication Date: 2019-05-03
• Subjects: Asymptotic distribution -- asymptotically distribution-free -- Balaam's design -- crossover design -- mixed carryover effect -- self carryover effect -- step-down approach; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5028505
• DOI: 10.1080/00949655.2019.1575381
• Languages: English
• ISSNs: 0094-9655

Abstract

ABSTRACT: The paper describes nonparametric approaches for comparing three-period, two-treatment, four-sequence crossover designs through testing the hypothesis that the treatments are interchangeable. The proposed approaches are based on a model which incorporates, along with the direct treatment effects, self and mixed carryover effects. Related asymptotic results are obtained. Comparisons among the designs are made numerically with respect to type I error rate and power of the tests considering compound symmetry and autoregressive covariance structures of the response variables. Lengths of the confidence intervals of the treatment differences are also used to make comparative study among the designs.