∫ ∫ Linear methods for regression and classification with functional data

Approximating the predic- tor and the response in finite dimensional functional spaces, we show the equivalence of the PLS regression with functional data and the finite

- Penalized functional linear model (Harezlak et.al (2007), Eilers and Marx (1996)).. Figure 1: Predicted and observed values for

In this context, various functional regression models (Ramsay and Silverman, 2005) have been proposed accord- ing to the nature of explanatory (or covariate) and response

The reader is referred to the forthcoming display (5) for an immediate illustration and to Mas, Pumo (2006) for a first article dealing with a functional autoregressive model

Keywords : Functional data, recursive kernel estimators, regression function, quadratic error, almost sure convergence, asymptotic normality.. MSC : 62G05,

Specifically, in functional regression analysis one intends to predict a random scalar variable (response variable) Y from a functional random variable (predic- tor) X (e.g.,

We suggest a nonparametric regression estimation approach which is to aggregate over space.That is, we are mainly concerned with kernel regression methods for functional random

To this aim we propose a parsimonious and adaptive decomposition of the coefficient function as a step function, and a model including a prior distribution that we name