51
5

Sequential multiple hypothesis testing in presence of control variables

Abstract

Suppose that at any stage of a statistical experiment a control variable XX that affects the distribution of the observed data YY at this stage can be used. The distribution of YY depends on some unknown parameter θ\theta, and we consider the problem of testing multiple hypotheses H1:θ=θ1H_1: \theta=\theta_1, H2:θ=θ2,...H_2: \theta=\theta_2, ..., Hk:θ=θkH_k: \theta=\theta_k allowing the data to be controlled by XX, in the following sequential context. The experiment starts with assigning a value X1X_1 to the control variable and observing Y1Y_1 as a response. After some analysis, another value X2X_2 for the control variable is chosen, and Y2Y_2 as a response is observed, etc. It is supposed that the experiment eventually stops, and at that moment a final decision in favor of one of the hypotheses H1,...H_1,..., HkH_k is to be taken. In this article, our aim is to characterize the structure of optimal sequential testing procedures based on data obtained from an experiment of this type in the case when the observations Y1,Y2,...,YnY_1, Y_2,..., Y_n are independent, given controls X1,X2,...,XnX_1,X_2,..., X_n, n=1,2,...n=1,2,....

View on arXiv
Comments on this paper