L1-Consistency of the kernel density estimators based on randomly right censored data
- 1 January 1990
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 19 (8) , 2853-2870
- https://doi.org/10.1080/03610929008830353
Abstract
Let [fcirc] PL plbe a kernel estimator of the density f(x) based on the product limit estimator of the distribution functionF(x) associated with the density f(x) In this paper convergence properties of are established. In particular, it is shown that J n (T)→completely as n→∞Also it is shown that this complete convergence is equivalent to weak and strong convergence of J n (T) to 0Keywords
This publication has 10 references indexed in Scilit:
- Strong Embedding of the Estimator of the Distribution Function under Random CensorshipThe Annals of Statistics, 1988
- Asymptotically Optimal Bandwidth Selection for Kernel Density Estimators from Randomly Right-Censored SamplesThe Annals of Statistics, 1987
- Some Asymptotic Properties of Kernel Estimators of a Density Function in Case of Censored DataThe Annals of Statistics, 1986
- A Histogram Estimator of the Hazard Rate with Censored DataThe Annals of Statistics, 1985
- A Buckley-James-type estimator for the mean with censored dataBiometrika, 1984
- Nonparametric density estimation from censored dataCommunications in Statistics - Theory and Methods, 1984
- The Equivalence of Weak, Strong and Complete Convergence in $L_1$ for Kernel Density EstimatesThe Annals of Statistics, 1983
- Strong consistency properties of nonparametric estimators for randomly censored data, II: Estimation of density and failure ratePeriodica Mathematica Hungarica, 1981
- Measure and IntegralPublished by Taylor & Francis ,1977
- Asymptotic Minimax Character of the Sample Distribution Function and of the Classical Multinomial EstimatorThe Annals of Mathematical Statistics, 1956