Local linear-$k$NN smoothing for semi-functional partial linear regression

Authors : Kedir Nassima Houda, Benchikh Tawfik, Naceri Amina, Fetitah Omar

Pages : 537-555

Doi:10.15672/hujms.1294382

View : 109 | Download : 88

Publication Date : 2024-04-23

Article Type : Research Paper

Abstract :The aim of this paper is to study a semi-functional partial linear regression model. The estimators are constructed by $k$-nearest neighbors local linear method. Some asymptotic results are established for an i.i.d sample under certain conditions, including asymptotic normality of the parametric component and the almost certain convergence (with rate) of the non-parametric component. Lastly, using cross-validation, the performances of our estimation method are presented on simulated data and on real data by comparing them with other known approaches for semi-functional partial linear regression models.
Keywords : Functional data analysis, partial linear regression, \\it k NN estimation, local linear estimation, asymptotic normality