A semiparametric hazard model with parametrized time but general covariate dependency is formulated and analyzed inside the framework of counting process theory. A profile likelihood principle is introduced for estimation of the parameters: the resulting estimator is $n^{1/2}$-consistent, asymptotically normal and achieves the semiparametric efficiency bound. An estimation procedure for the nonparametric part is also given and its asymptotic properties are derived. We provide an application to mortality data.
Publié le : 1998-02-14
Classification:
Counting process theory,
kernel estimation,
predictability,
semiparametric survival analysis,
62G05,
62M09
@article{1030563983,
author = {Nielsen, Jens P. and Linton, Oliver and Bickel, Peter J.},
title = {On a semiparametric survival model with flexible covariate effect},
journal = {Ann. Statist.},
volume = {26},
number = {3},
year = {1998},
pages = { 215-241},
language = {en},
url = {http://dml.mathdoc.fr/item/1030563983}
}
Nielsen, Jens P.; Linton, Oliver; Bickel, Peter J. On a semiparametric survival model with flexible covariate effect. Ann. Statist., Tome 26 (1998) no. 3, pp. 215-241. http://gdmltest.u-ga.fr/item/1030563983/