The problem of selecting the better Bernoulli treatment using a matched samples design is considered in the framework of the indifference-zone approach. A single-stage procedure is proposed and its properties are studied. Tables of sample sizes for implementing the proposed procedure are given. A comparison is made with an independent samples design and the associated Sobel-Huyett selection procedure.
|Journal||Journal of the Royal Statistical Society, Series B (Methodological)|
|State||Published - 1980|