A self-designing rule for clinical trials with arbitrary response variables

For testing both one-sided and two-sided hypotheses concerning several trea tment arms in group sequentially performed clinical trials with arbitrary o utcome variables, a general method is considered that allows one to complet ely self-design a study. All information available prior to a stage is used for estimating the sample size and the weight for the next step. In "using up" the variance, the test statistic is built in a bounded finite but rand om number of stages to test just once the rejection of the null hypothesis. (C) Elsevier Science Inc. 2001.