Quadratic error of the conditional hazard function in the local linear estimation for functional data

Abstract

In this paper we investigate the asymptotic mean square error and the rates of convergence of the estimator based on the local linear method of the conditional hazard function. Under some general conditions, the expressions of the bias and variance are given. The efficiency of our estimator is evaluated through a simulation study. We proved, theoretically and on the scope of a simulation study, that our proposed estimator has better performance than the estimator based on the standard kernel method.

Keywords: Nonparametric local linear estimation, conditional hazard function, functional variable, mean squared error.

AMS 2010 Mathematics Subject Classification: 62G05, 62G20