Conditional residual lifetime under random censorship

Abstract

In this paper we study the nonparametric estimation of two important functionals of the residual lifetime beyond some fixed or random timepoint: the mean and any quantile. The observations are subject to random right censoring and covariate information is taken into account. In this generality we obtain asymptotic representations and asymptotic normality results for the proposed estimators. Several results in the literature are thereby generalized to the regression context.