Title: Bandwidth selection of recursive nonparametric relative regression for independent functional data
Authors: Yousri Slaoui - University of Poitiers (France) [presenting]
Abstract: New kernel regression estimators are proposed based on the minimization of the mean squared relative error. We study the properties of the proposed recursive estimators and compare them with the recursive estimators based on the minimization of the mean squared error. It turns out that, with an adequate choice of the parameters, the proposed estimators outperformed the recursive estimators based on the minimization of the mean squared error in some specific situations as the presence of outliers or when the response of the model is usually positive. We corroborate these theoretical results through a real chemometric dataset.