A New Confidence Interval for Odds Ratio

TL;DR

This paper introduces a new confidence interval for the odds ratio that guarantees at least the nominal coverage probability, addressing limitations of existing asymptotic methods in medical research.

Contribution

A novel confidence interval for the odds ratio is proposed, requiring only sample sizes and sample odds ratio, with improved coverage probability.

Findings

Coverage probability is at least the nominal level.

Requires only sample sizes and sample odds ratio.

Addresses poor coverage of existing intervals.

Abstract

We consider the problem of interval estimation of the odds ratio. An asymptotic confidence interval is widely applied in medical research. Unfortunately that confidence interval has a poor coverage probability: it is significantly smaller than the nominal confidence level. In this paper a new confidence interval is proposed. The construction needs only information on sample sizes and sample odds ratio. The coverage probability of the proposed confidence interval is at least the nominal confidence level.