A central limit theorem for the integrated square error of the kernel density estimators with randomly censored data (Q1314484)

To estimate the density in a random censorship model a kernel estimator is used where instead of the empirical distribution function the Kaplan- Meier estimator is applied. The integrated square error is used to describe the accuracy of the density estimator. To investigate the asymptotic behaviour of this error a central limit theorem is established which can be applied to construct goodness of fit tests and to the Hellinger distance parametric inference.