Interval Censored Survival Data : A Review of Recent

We review estimation in interval censoring models, including nonparametric estimation of a distribution function and estimation of regression models. In the non-parametric setting, we describe computational procedures and asymptotic properties of the nonparametric maximum likelihood estimators. In the regression setting, we focus on the proportional hazards, the proportional odds and the accelerated failure time semiparametric regression models. Particular emphasis is given to calculation of the Fisher information for the regression parameters. We also discuss computation of the regression parameter estimators via proole likelihood or maximization of the semi-parametric likelihood, distributional results for the maximum likelihood estimators, and estimation of (asymptotic) variances. Some further problems and open questions are also reviewed. 1 0. Outline.