Multi-Layer Neural Network with functional inputs: an inverse regression approach
Louis Ferré, Nathalie Villa · 2005
Functional data analysis is a growing research eld since more and more pratical applications involve functional data. In this paper, we focus on the problem of regression and classication with functional predictors: the model suggested combines an ecient dimension reduction procedure (functional SIR, rst introduced by Ferré and Yao (2003)), for which we give a regularized version, and the accuracy of a neural network. The con-sistency of the model is proved and the method is successfully confronted to real life data.