Francesco Bravo (University of York)

Semiparametric quantile regressions with random censoring

Abstract

This paper considers estimation in semiparametric quantile regression models when the response variable is subject to random censoring. The paper considers both the case of independent and dependent censoring and proposes three iterative estimators based on inverse probability weighting, where the weights are estimated from the censoring distribution using the Kaplan-Meier, a fully parametric and the conditional Kaplan-Meier estimator. The paper also proposes a computationally simple resampling technique that can be used to approximate the finite sample distribution of the finite dimensional parameter estimator. Monte Carlo simulations show that the proposed estimators have good finite sample properties. Finally the paper contains a real data application.