We consider bias-reduced estimation of the extreme value index in conditional Pareto-type models with random covariates when the response variable is subject to random right censoring. The bias-correction is obtained by fitting the extended Pareto distribution locally to the relative excesses over a high threshold using the maximum likelihood method. Consistency and asymptotic normality of the estimators are established under suitable assumptions. The finite sample behaviour is illustrated with a small simulation experiment and the method is applied to AIDS survival data.
Publié le : 2018-06-29
Classification:
Pareto-type model,
random covariate,
random right censoring,
local estimation,
bias-reduction,
[MATH.MATH-ST]Mathematics [math]/Statistics [math.ST]
@article{hal-01826112,
author = {Goegebeur, Yuri and Guillou, Armelle and Qin, Jing},
title = {Bias-corrected estimation for conditional Pareto-type distributions with random right censoring},
journal = {HAL},
volume = {2018},
number = {0},
year = {2018},
language = {en},
url = {http://dml.mathdoc.fr/item/hal-01826112}
}
Goegebeur, Yuri; Guillou, Armelle; Qin, Jing. Bias-corrected estimation for conditional Pareto-type distributions with random right censoring. HAL, Tome 2018 (2018) no. 0, . http://gdmltest.u-ga.fr/item/hal-01826112/