New $M$-estimators in semi-parametric regression with errors in variables
Abstract
In the regression model with errors in variables, we observe i.i.d. copies of satisfying and involving independent and unobserved random variables plus a regression function , known up to a finite dimensional . The common densities of the 's and of the 's are unknown, whereas the distribution of is completely known. We aim at estimating the parameter by using the observations . We propose an estimation procedure based on the least square criterion where is a weight function to be chosen. We propose an estimator and derive an upper bound for its risk that depends on the smoothness of the errors density and on the smoothness properties of . Furthermore, we give sufficient conditions that ensure that the parametric rate of convergence is achieved. We provide practical recipes for the choice of in the case of nonlinear regression functions which are smooth on pieces allowing to gain in the order of the rate of convergence, up to the parametric rate in some cases. We also consider extensions of the estimation procedure, in particular, when a choice of depending on would be more appropriate.
Cite
@article{arxiv.math/0511105,
title = {New $M$-estimators in semi-parametric regression with errors in variables},
author = {Cristina Butucea and Marie-Luce Taupin},
journal= {arXiv preprint arXiv:math/0511105},
year = {2009}
}
Comments
Published in at http://dx.doi.org/10.1214/07-AIHP107 the Annales de l'Institut Henri Poincar\'e - Probabilit\'es et Statistiques (http://www.imstat.org/aihp/) by the Institute of Mathematical Statistics (http://www.imstat.org)