Geodesic
Estimation of the extreme survival probabilities from censored data
Buletinul Academiei de Ştiinţe a Republicii Moldova. Matematica, no. 1 (2014), pp. 33-62

Voir la notice de l'article provenant de la source Math-Net.Ru

The Kaplan–Meier nonparametric estimator has become a standard tool for estimating a survival time distribution in a right censoring schema. However, if the censoring rate is high, this estimator does not provide a reliable estimation of the extreme survival probabilities. In this paper we propose to combine the nonparametric Kaplan–Meier estimator and a parametric-based model into one construction. The idea is to fit the tail of the survival function with a parametric model while for the remaining to use the Kaplan–Meier estimator. A procedure for the automatic choice of the location of the tail based on a goodness-of-fit test is proposed. This technique allows us to improve the estimation of the survival probabilities in the mid and long term. We perform numerical simulations which confirm the advantage of the proposed method. 
@article{BASM_2014_1_a3,
     author = {Ion Grama and Jean-Marie Tricot and Jean-Fran\c{c}ois Petiot},
     title = {Estimation of the extreme survival probabilities from censored data},
     journal = {Buletinul Academiei de \c{S}tiin\c{t}e a Republicii Moldova. Matematica},
     pages = {33--62},
     publisher = {mathdoc},
     number = {1},
     year = {2014},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/BASM_2014_1_a3/}
}
TY  - JOUR
AU  - Ion Grama
AU  - Jean-Marie Tricot
AU  - Jean-François Petiot
TI  - Estimation of the extreme survival probabilities from censored data
JO  - Buletinul Academiei de Ştiinţe a Republicii Moldova. Matematica
PY  - 2014
SP  - 33
EP  - 62
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/BASM_2014_1_a3/
LA  - en
ID  - BASM_2014_1_a3
ER  - 
%0 Journal Article
%A Ion Grama
%A Jean-Marie Tricot
%A Jean-François Petiot
%T Estimation of the extreme survival probabilities from censored data
%J Buletinul Academiei de Ştiinţe a Republicii Moldova. Matematica
%D 2014
%P 33-62
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/item/BASM_2014_1_a3/
%G en
%F BASM_2014_1_a3
Ion Grama; Jean-Marie Tricot; Jean-François Petiot. Estimation of the extreme survival probabilities from censored data. Buletinul Academiei de Ştiinţe a Republicii Moldova. Matematica, no. 1 (2014), pp. 33-62. http://geodesic.mathdoc.fr/item/BASM_2014_1_a3/