Goodness-of-Fit Tests in Logistic Regression with Continuous Covariates

Abstract

Goodness-of-fit tests are used to assess model adequacy. There are a number of ways in which a fitted model can be inadequate (Xie et al. (2008)). For instance, the linear systematic component of the model may be incorrectly specified, a covariate may not be specified in the appropriate functional form, some important covariates may have been omitted from the model, or the link function may be misspecified. All these model misspecifications could affect consistency of the coefficient estimation, and can lead to biased estimates of treatment effects (Gail et al. (1988); Hauck et al. (1991)).