In this paper we are concerned with the problem of comparing adjacent ordered treatments in a one-way layout where survival data are subject to random right censorship. Multiple testing procedures based on two-sample statistics, each comparing an individual treatment with the previous one, are proposed for determining the pattern of the treatment effects. The two-sample statistics under consideration are weighted logrank statistics (Fleming and Harrington, 1991, Counting Process and Survival Analysis. Wiley, New York) and weighted Kaplan-Meier statistics (Pepe and Fleming, 1989. Biometrics 45, 497-507; 1991. J. Roy. Statist. Soc. B 53, 341-352). An illustrated numerical example is reported. Finally, the comparative results of a Monte Carlo error rate and power study for small sample sizes are presented.
All Science Journal Classification (ASJC) codes