PARAMETER-ESTIMATION FROM INCOMPLETE DATA IN BINOMIAL REGRESSION WHENTHE MISSING DATA MECHANISM IS NONIGNORABLE

We propose a method for estimating parameters in binomial regression m odels when the response variable is missing and the missing data mecha nism is nonignorable. We assume throughout that the covariates are ful ly observed. Using a legit model for the missing data mechanism, we sh ow how parameter estimation can be accomplished using the EM algorithm by the method of weights proposed in Ibrahim (1990, Journal of the Am erican Statistical Association 85, 765-769). An example from the Six C ities Study (Ware et al., 1984, American Review of Respiratory Disease s 129, 366-374) is presented to illustrate the method.