On semiparametric random censorship models (Q1298702)

In this interesting paper, a new semiparametric model for data under random censorship is introduced and studied in detail. The distribution of the observed \((Z, \delta)\), where \(Z = \min(X,Y)\) and \(\delta = I\{ X \leq Y\}\) is uniquely determined by the distribution of \(Z\) and the conditional probability \(m(z)\), that given \(Z = z\) the indicator \(\delta\) equals 1. A general class of parametric models for \(m\) is introduced, which contains the Koziol-Green model as a special case. The paper provides the asymptotic theory for estimators of the parametric part, \(m\), and (functional) distributional convergence for the associated empirical and cumulative hazard processes.