On the asymptotic distribution of maximum likelihood estimators with doubly censored data

Abstract

A method of estimating lifetime parameters from doubly censored data is given on the basis of the maximum likelihood principle. Consistency and asymptotic normality of the proposed estimator is established. When the observation window is defined by the interval of time between the first and the pth events in a homogeneous Poisson process, asymptotic variances and efficiencies are analyzed assuming exponential lifetime distribution.