We present an efficient network algorithm for generating exact permutational distributions for linear rank tests defined on stratified 2 x c contingency tables. The algorithm can evaluate exact one and two sided p-values, and compute exact confidence intervals for trend parameters arising from certain log linear and logistic models embedded in these contingency tables. It is especially efficient for highly imbalanced categorical data, a situation where the asymptotic theory is unreliable. Part of the algorithm can be adapted to evaluating the conditional maximum likelihood and its derivatives for the logistic regression model, with grouped data. We illustrate the techniques with an analysis of two data sets the leukemia data on the Hiroshima atomic bomb survivors, and data from a clinical trial of bone marrow transplant.
This article is from 'Computing Science and Statistics: Proceedings of the Symposium on the Interface Critical Applications of Scientific Computing: Biology, Engineering, Medicine, Speech Held in Seattle, Washington on 21-24 April 1991,' ADA252938, p200-207.