Hazard regression with non compactly supported bases

Elodie Brunel (1) , Fabienne Comte (2)

1 IMAG - Institut Montpelliérain Alexander Grothendieck
2 MAP5 - UMR 8145 - Mathématiques Appliquées Paris 5

In this paper, we consider the problem of nonparametric hazard rate estimation in presence of right-censored observations. We provide a generalized risk bound for a regression type nonparametric estimator of the hazard function of interest. Under adequate integrability conditions, our bound is a generalization to non necessarily compactly supported bases, of strategies which were specific to compact support of estimation. We show that it encompasses those previous compact-support results. We discuss the model selection method which comes out from the new terms of the risk bounds, and compare the performance of the new estimator to previous ones, when using a non compact Laguerre basis. A real data example is also presented.

Mots clés: Hazard rate. Laguerre basis. Least squares regression. Nonparametric estimation. Projection estimator.

Domaines: Statistiques [math.ST]

Elodie Brunel, Fabienne Comte. Hazard regression with non compactly supported bases. Canadian Journal of Statistics, 2021, 49 (4), pp.1273-1297. ⟨10.1002/cjs.11619⟩. ⟨hal-02615228⟩