作者: Colin B. Begg , Boris Iglewicz
DOI: 10.2307/2530497
关键词: Function (engineering) 、 Programmable calculator 、 Biometrics 、 Minification 、 Mathematical optimization 、 Computer science 、 Clinical trial 、 Linear model 、 Variance (accounting) 、 Research design
摘要: A dynamic treatment allocation procedure is proposed for clinical trials which require balancing across several prognostic factors. The decision based on the minimization of a function an easily evaluated approximation to variance effect in linear model relating outcome variable chosen factors and selected interactions. By use simulations, shown be superior ad hoc procedures by Pocock Simon (1975, Biometrics 31, 103-115), over variety reasonable experimental situations. It that it feasible evaluate hand calculations extremely easy if small programmable calculator available. Practical problems implementation are discussed with special reference multi-institutional trials.