On Logit Confidence Intervals for the Odds Ratio with Small Samples

作者：Alan Agresti · 发表于：Biometrics · 年份：1999 · DOI：10.1111/j.0006-341x.1999.00597.x · 被引用次数：115 · 研究领域：Statistical Methods in Clinical Trials、Meta-analysis and systematic reviews、Statistical Methods and Bayesian Inference

Unless the true association is very strong, simple large-sample confidence intervals for the odds ratio based on the delta method perform well even for small samples. Such intervals include the Woolf logit interval and the related Gart interval based on adding .5 before computing the log odds ratio estimate and its standard error. The Gart interval smooths the observed counts toward the model of equiprobability, but one obtains better coverage probabilities by smoothing toward the independence model and by extending the interval in the appropriate direction when a cell count is zero.