A1346
Title: Finite-sample improvements in nonparametric functional regression through weighted pseudo-metrics
Authors: Kwo Lik Lax Chan - Universita degli Studi del Piemonte Orientale (Italy) [presenting]
Laurent Delsol - University of Orleans (France)
Aldo Goia - University of Eastern Piedmont Amedeo Avogadro (Italy)
Abstract: One of the main problems in functional data analysis is the selection of pseudo-metric as a distance measure between curves; in particular, in the nonparametric regression context, it has a direct impact as it captures the information contained in the explanatory curve by extracting the informative features of the explanatory curve. The idea of weighted pseudo-metric is introduced, implemented, and discussed. Performances of the choices of pseudo-metric used are evaluated by means of a Monte Carlo Study.
