Optimal and fast confidence intervals for hypergeometric successes

TL;DR

This paper introduces an efficient method for calculating exact confidence intervals for hypergeometric success parameters, optimizing interval size and computational speed, and provides an R package implementation.

Contribution

It presents a novel approach to invert acceptance intervals for hypergeometric confidence intervals, achieving minimal average size and fast computation.

Findings

Intervals are smaller than existing methods.

Computation time is significantly reduced.

The method is implemented in the R package hyperMCI.

Abstract

We present an efficient method of calculating exact confidence intervals for the hypergeometric parameter representing the number of "successes," or "special items," in the population. The method inverts minimum-width acceptance intervals after shifting them to make their endpoints nondecreasing while preserving their level. The resulting set of confidence intervals achieves minimum possible average size, and even in comparison with confidence sets not required to be intervals it attains the minimum possible cardinality most of the time, and always within $1$. The method compares favorably with existing methods not only in the size of the intervals but also in the time required to compute them. The available \textsf{R} package \texttt{hyperMCI} implements the proposed method.