Additive Reversed Hazard Rates Models

SYNOPTIC ABSTRACT

The additive reversed hazards model relates the conditional reversed hazard function of the lifetime linearly to the covariates. It describes the association between the lifetime and covariates in terms of risk difference. In the present work, we introduce an additive reversed hazards model for modeling and analysis of lifetime in the presence of covariates under left censoring. We develop a closed form semiparametric estimator of the regression parameter. We also provide a Breslow-type estimator for a cumulative baseline reversed hazard function. Asymptotic properties of the estimators are studied. Simulation studies are conducted to assess the finite sample properties of the estimators. Finally, we apply the proposed model to real-life data.

Key Words and Phrases::

• reversed hazard function
• additive model
• regression model
• left censoring