×

Asymptotic results for estimators in restricted randomization designs. (English) Zbl 0585.62131

In clinical trials one may occasionally want to manipulate the design during the course of the study. The object may be to achieve a balance between the number of individuals in each treatment group, or to ensure that more individuals receive the better treatment.
A simple situation with a dichotomous outcome is considered, and a result about consistency and asymptotic normality is presented. The results are applied to two special cases. This paper contains nice applications of martingale central limit theory and demonstrates its usefulness for inferential statistics.
Reviewer: G.Kersting

MSC:

62L05 Sequential statistical design
62P10 Applications of statistics to biology and medical sciences; meta analysis
60G42 Martingales with discrete parameter
PDFBibTeX XMLCite