Extension of the Wilcoxon-Mann-Whitney Test to Samples Censored at the Same Fixed Point

Abstract

A two-sample non-parametric significance test for censored samples is presented. The null hypothesis considered is that the two samples are from the same population. Tables of lower 5% and 1% points of the test statistic are given for both sample sizes ≦8 and various degrees of censoring. The asymptotic distribution of the test statistic under the null hypothesis is derived and numerical comparisons are given which indicate the asymptotic theory is adequate outside the range of the table providing no more than about 75% of the total observations of the two samples are censored. The test is shown to be consistnet against an alternative which, for example, includes the case of sampling from two normal populations with different means.