Multilayer Perceptron with Functional Inputs: an Inverse Regression Approach

Louis Ferré (1) , Nathalie Villa (1)

1 GRIMM - Groupe de Recherche en Mathématiques et Informatique du Mirail

Abstract. Functional data analysis is a growing research field as more and more practical applications involve functional data. In this paper, we focus on the problem of regression and classification with functional predictors: the model suggested combines an efficient dimension reduction procedure [functional sliced inverse regression, first introduced by Ferré & Yao (Statistics, 37, 2003, 475)], for which we give a regularized version, with the accuracy of a neural network. Some consistency results are given and the method is successfully confronted to real-life data.

Domaines: Statistiques [math.ST]

Louis Ferré, Nathalie Villa. Multilayer Perceptron with Functional Inputs: an Inverse Regression Approach. Scandinavian Journal of Statistics, 2006, 33 (4), pp.807-823. ⟨10.1111/j.1467-9469.2006.00496.x⟩. ⟨hal-00144144⟩