RANDOM ALLOCATION DESIGN FOR CONTROL OF SELECTION BIAS IN SEQUENTIAL EXPERIMENTS
成果类型：
Article
署名作者：
WEI, LJ
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
DOI：
10.2307/2335280
发表日期：
1978
页码：
7984
关键词：
摘要：
In comparing 2 treatments, suppose that eligible [human] subjects come to the experiment sequentially and must be treated immediately. It was assumed that the experimenter selects 2n subjects and treats n subjects with each treatment. If, during the experiment, the experimenter knows or guesses which treatment is to be administered next, one may bias the experiment by subject selection. The problem is how to control this bias through choice of design. A design criterion is derived. Under this criterion the random allocation design is superior to the truncated binomial design which has been shown to be the minimax one by Blackwell and Hodges (1957) under their design criterion.