A Nonparametric Test for the Comparison of Multidimensional Outcomes

Abstract

A Nonparametric Test for the Comparison of Multidimensional Outcomes Kai Fun Yu Department of Mathematical Statistics, Tsinghua University, China ?: A nonparametric test based on the rank-sums is developed to improve the test procedures by O’Brien (1984) and Huang et al. (2005) when the differences between the comparison groups in each outcome variable may be in different directions. The proposed test statistic which takes the maximum of the individual rank-sum statistic controls the type I error rate and maintains satisfactory power regardless of the direction of the differences. Simulation studies show the proposed test to be of higher power than other tests in a certain alternative parameter space of interest. Some real examples are presented. On these examples, the proposed test yields more satisfactory results.