Аннотация:Two statistics for testing the equality of survival distributions in the presence of arbitarry right censorship, the log rank test statistic and the modified Wilcoxon test statictic, are shown to differ only in the choice of weights which are funcitons of the number of individuals at risk at the time of each death. A related statistic is proposed which may be more powerful across a range of alternatives than are the log rank statistic and the modified Wilcoxon statistic.
