Confidence interval for relative difference under inverse sampling

Abstract

Inverse sampling is one of the hot topics in the continuous sampling process. It is considered to be a more appropriate sampling scheme than the usual binomial sampling scheme when the subjects arrive sequentially and the underlying response of interest is acute. In this paper, we propose five confidence interval estimators for the relative difference in 2 2 × table under inverse sample. The performance of the proposed methods is evaluated with respect to their coverage probability and expected widths. Our empirical results show that the confidence intervals based on the score statistic produce similar coverage probabilities and smaller confidence interval widths.