A General Semiparametric Approach to Inference with Marker-Dependent Hazard Rate Models

We examine a new general class of hazard rate models for duration data, containing a parametric and a nonparametric component. Both can be a mix of atime effect and possibly time-dependent covariate effects. A number of well-known models are special cases. In a counting process framework, a general profile like-lihood estimator is developed and the parametric component of the model isshown to be asymptotically normal and efficient. Finite sample properties areinvestigated in simulations. The estimator is applied to investigate the long-runrelationship between birth weight and later-life mortality