Continuous auxiliary covariate in additive hazards regression for survival data

Abstract

Listen

Translate article

This paper considers the additive hazards regression analysis by utilizing continuous auxiliary covariate information to improve the efficiency of the statistical inference when the primary covariate is ascertained only for a randomly selected subsample. The authors construct a martingale-based estimating equation for the regression parameter and establish the asymptotic consistency and normality of the resultant estimators. Simulation study shows that the proposed method can greatly improve the efficiency compared with the estimator which discards the auxiliary covariate information in a variety of settings. A real example is also provided as an illustration.