Modified Kernel Quantile Estimator under Censoring

摘要

Based on right censored data from a lifetime distribution F sub 0, a modification of the kernel quantile estimator is proposed. The advantage of this estimator is that the data play a role in the degree of smoothing of the estimator while retaining the desireable features of the kernel estimator. Convergence in probability and almost sure convergence of the estimator are discussed. Some examples are given which illustrate the differences between this modified estimator and the fixed-bandwidth kernel quantile estimator for randomly right-censored data.