PERMUTATION TESTS FOR COMMON LOCATIONS AMONG SAMPLES WITH UNEQUAL VARIANCES

In completely randomized experimental designs where population varianc es are equal under the null hypothesis, it is not uncommon to have mul tiplicative treatment effects that produce unequal variances under the alternative hypothesis. Permutation procedures are presented to test for (a) median location and scale shifts, (b) scale shifts only, and ( c) mean location shifts only. Corresponding multivariate extensions ar e provided. Location-shift power comparisons between the parametric Ba rtlett-Nanda-Pillai trace test and three alternative multivariate perm utation tests for five bivariate distributions are included.