Abstract
We show that a nonparametric estimator of a regression function, obtained as solution of a specific regularization problem is the bestlinear unbiased predictor in some nonparametric mixed effect model. Since this estimator is
intractable from a numerical point of view, we propose a tight approximation of it easy and fast to implement. This second estimator achieves the usual optimal rate of convergence of the mean integrated squared error over a Sobolev class both for equispaced and non equispaced design. Numerical experiments are presented both on simulated and ERP real data.
Anno
2003
Autori IAC
Tipo pubblicazione
Altri Autori
Angelini C., De Canditiis D., Leblanc F.
Editore
Academic Press.
Rivista
Journal of Multivariate Analysis (Print)