Pairwise comparisons of the means of skewed data

Abstract

This paper considers the problem of pairwise comparisons of the means of skewed data with special emphasis on log-normal distributions. Bootstrap procedures are proposed to approximate the sampling distribution of the pivotal statistic generating the simultaneous confidence sets of the pairwise differences. The operating characteristics of the proposed methods are compared with the widely used Bonferroni pairwise procedure in a simulation study. A clinical example is used to illustrate the proposed procedures.