Likelihood process in parametric model of censored data with staggered entry-asymptotic properties and applications (Q1116581)

A parametric model for survival data with staggered entry in a clinical trial is specified. Under some regularity conditions, weak convergence and consistency of the score process and the information process are established, using a martingale limit theorem and Lenglart's inequality. Its application yields the asymptotic properties of the maximum likelihood estimator process and the local asymptotic normality of the parametric family in the D-space with Skorokhod topology. Similar results for stopped log-likelihood processes are also given.