An O'Brien-Fleming sequential trial for comparing three treatments

摘要: We consider a sequential procedure for comparing three treatments with the goal of ultimately selecting best treatment. This starts test to detect an overall treatment difference and eliminates apparently inferior if this rejects equality treatments. It then proceeds remaining two base these tests on stopping boundaries popularized by O'Brien Fleming. Our is similar in structure that used Siegmund conjunction modified repeated significance tests. compare performances procedures via simulation experiment. derive analytic approximations error probability, power expected sample size our procedure, which we simulated values. Furthermore, propose modification comparison standard experimental