Bias of Odds Ratio Estimate in Fisher's Exact Test

Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine

ABSTRACT Objectives The odds ratio estimate in Fisher's exact test can overestimate the parameter. A simple computer simulation can easily reveal the positive bias of the odds ratio estimate from Fisher's exact test. Bootstrap can facilitate bias correction for the odds ratio estimate. mpr70076-sec-0001 Methods The bias can be estimated, using bootstrap samples and the original sample to approximate the expectation of the odds ratio estimator and the true parameter value—their difference is the bias. Here, the bias is computed from the underlying distribution, conditional on the exclusion of zero cells in sampling, to avoid the infinite expectation. mpr70076-sec-0002 Results A study of depression is used to demonstrate how to use bootstrap to correct the bias in an odds ratio estimate based on Fisher's exact test. mpr70076-sec-0003 Conclusions Bootstrapping can easily estimate and correct the bias of an odds ratio estimate in Fisher's exact test. The results suggest that bootstrapping is sensitive enough to detect even a small bias. mpr70076-sec-0004