论文摘要：With censored failure time data, we consider the estimation procedure and statistical inferences for time-varying coefficients under the additive hazards model. We propose to maximize the local likelihood function for the estimation of the time-varying coefficients and establish the asymptotic normality of the estimators. The boundary effect is examined and the large sample properties for the modified estimators at the boundary points are derived. The variance and bias of the coefficient estimator can be consistently estimated. We conduct simulation studies to examine the performance of the proposed method for finite samples. As an illustration, we apply the proposed method to a real data example from a bone marrow transplantation study.
论文题目: Estimation and Inference of Varying-Coefficient Additive Hazards Model
论文作者: Li, H. Yin, G. and Zhou, Y
发表刊物: Canadian Journal of Statistics, 35（2），321-337
附件下载：
