|
Abstract:
|
Random controlled experiment is the "gold standard" of the study designs which focus at treatment comparisons in health related field[7]. The validity of statistical inference only underwrite from proper randomization process. In a random controlled experiment, subjects are randomly assigned to one of the treatment groups. Several approach has been studies to restrict treatment allocation process to ensure the validity of statistical inference in those studies, such as complete randomization, semi-randomization and non-randomization method. The idea of treatment allocation based on ordered statistic is inspired by ranked set sampling. Ranked set sampling (RSS) was first introduced by McIntyre [12]. The principle of RSS is to select sample systematically from a population. It was originally introduced as an efficient alternative to simple random sampling for estimating the field of pastures. We introduce a method based on ranked auxiliary variables for treatment allocation in crossover designs using Latin square models.
|
