Abstract Frequently the auditor wishes to combine the Stringer bounds for several populations that have been sampled independently. Relatively little is known at present about how to combine the Stringer bounds for the different populations when varying sampling fractions have been employed. This paper investigates by means of a simulation study five plausible methods of combining Stringer bounds under such circumstances. The investigation focuses on the coverage and relative tightness of the combined bounds obtained with these five methods. The results indicate that one of the methods -- the global combined bound -- almost always leads to the tightest bound, and always provides high coverage. The results further suggest that the superiority of this method increases as more populations are to be combined.
Neter et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: