ABSTRACT
Binary response data, measured with covariates, are often modeled by assuming that the probability of a positive response, after a suitable transformation, is linear in the covariates. This transformation, usually known as link function, connects the probability to a linear predictor and is normally assumed to be a known function, for instance the inverse cumulative function of a certain distribution. McCullagh and Neider (1989) provides a comprehensive discussion of the use of link functions to model the relation between a binary response and linear covariates.
