Application of generalized linear models to the analysis of toxicity test data

Generalized linear models give a unified approach to the performance of regression analysis of dichotomous, count or continuous data. This paper studies binomial, negative binomial and gamma regression models and gives a detailed description of inference procedures based on them. In particular, the maximum likelihood procedure is described for a logistic function (binomial regression) or a log-linear regression model (negative binomial and gamma regression). The process of model fitting and evaluation is illustrated by examples referring to the determination of endpoints in acute and chronic toxicity tests.