(Received January 4, 1996; revised August 6, 1996)
Abstract. In this paper a simple way to obtain empirical likelihood type confidence intervals for the mean under random censorship is suggested. An extension to the more general case where the functional of interest is an M-functional is discussed and the proposed technique is used to construct confidence intervals for quantiles. The results of a simulation study carried out to assess the accuracy of these inferential procedures are also given.
Key words and phrases: Censored data, empirical likelihood, Kaplan-Meier estimator, M-estimator, M-functional, profile likelihood.