[Estimation des probabilités de survie par lʼajustement sur la queue dʼun modèle de Cox]
Dans le contexte dʼun modèle dʼanalyse de survie avec données censurées et en présence dʼune covariable explicative, on veut prédire la probabilité de survie au-delà de la plus grande observation. Un modèle de Cox dont la fonction de risque de base est constante, est ajusté sur la queue de la distribution de la durée de vie. Sous certaines conditions de régularité, on démontre la convergence des estimateurs des paramètres du modèle.
Within the framework of a survival analysis model with censored life time data and an explanatory covariate, our goal is to predict the survival probability beyond the largest observed time. A Cox model with a constant underlying hazard function is proposed to adjust the tail of the life time distribution. Under some regularity conditions, we prove that the parameter estimators are convergent.
Accepté le :
Publié le :
@article{CRMATH_2011__349_13-14_807_0,
author = {Grama, Ion and Tricot, Jean-Marie and Petiot, Jean-Fran\c{c}ois},
title = {Estimation of the survival probabilities by adjusting a {Cox} model to the tail},
journal = {Comptes Rendus. Math\'ematique},
pages = {807--811},
publisher = {Elsevier},
volume = {349},
number = {13-14},
year = {2011},
doi = {10.1016/j.crma.2011.06.011},
language = {en},
url = {http://www.numdam.org/articles/10.1016/j.crma.2011.06.011/}
}
TY - JOUR AU - Grama, Ion AU - Tricot, Jean-Marie AU - Petiot, Jean-François TI - Estimation of the survival probabilities by adjusting a Cox model to the tail JO - Comptes Rendus. Mathématique PY - 2011 SP - 807 EP - 811 VL - 349 IS - 13-14 PB - Elsevier UR - http://www.numdam.org/articles/10.1016/j.crma.2011.06.011/ DO - 10.1016/j.crma.2011.06.011 LA - en ID - CRMATH_2011__349_13-14_807_0 ER -
%0 Journal Article %A Grama, Ion %A Tricot, Jean-Marie %A Petiot, Jean-François %T Estimation of the survival probabilities by adjusting a Cox model to the tail %J Comptes Rendus. Mathématique %D 2011 %P 807-811 %V 349 %N 13-14 %I Elsevier %U http://www.numdam.org/articles/10.1016/j.crma.2011.06.011/ %R 10.1016/j.crma.2011.06.011 %G en %F CRMATH_2011__349_13-14_807_0
Grama, Ion; Tricot, Jean-Marie; Petiot, Jean-François. Estimation of the survival probabilities by adjusting a Cox model to the tail. Comptes Rendus. Mathématique, Tome 349 (2011) no. 13-14, pp. 807-811. doi : 10.1016/j.crma.2011.06.011. http://www.numdam.org/articles/10.1016/j.crma.2011.06.011/
[1] Asymptotic theory for the Cox model with missing time dependent covariate, Ann. Statist., Volume 34 (2006), pp. 903-924
[2] Pareto approximation of the tail by local exponential modeling, Bull. Acad. Sci. Moldova, Volume 1 (2007) no. 53, pp. 1-22
[3] Statistics of extremes by oracle estimation, Ann. Statist., Volume 36 (2008), pp. 1619-1648
[4] Best attainable rates of convergence for estimates of parameters of regular variation, Ann. Statist., Volume 12 (1984), pp. 1079-1084
[5] Adaptive estimates of regular variation, Ann. Statist., Volume 13 (1985), pp. 331-341
[6] The Statistical Analysis of Failure Time Data, Wiley, 2002
[7] Survival Analysis: Techniques for Censored and Truncated Data, Springer, 2003
Cité par Sources :
