Goodness-of-fit tests for randomly censored data

Abstract

Many of the proposed tailor-made goodness-of-fit tests for complete data can be represented as a functional of some V-empirical processes. Here we present one of the possible approaches for the adaptation of such tests for randomly right censored data and show general results about the asymptotic properties of newly adapted statistics. Theoretical results are demonstrated on several classes of goodness-of-fit tests.