Estimation of the Functional Linear Regression with Smoothing Splines
We consider functional linear regression where a real variable Ydepends on a functional variable X. The functional coeficient of the model is estimated by means of smoothing splines. We derive the rates of convergence with respect to the semi-norm induced by the covariance operator of X, which comes to evaluate the error of prediction. These rates, which essentially depend on the smoothness of the function parameter and on the structure of the predictor, are shown to be optimal over a large class of functions parameters and distributions of the predictor.
KeywordsFunction Parameter Covariance Operator Functional Data Analysis Functional Principal Component Natural Spline
Unable to display preview. Download preview PDF.
- Crambes, C., Kneip, A. and Sarda, P.: Smoothing splines estimators for functional linear regression. To appear. (2007).Google Scholar
- Eubank, R.L.: Spline smoothing and nonparametric regression. Marcel Dekker. (1988).Google Scholar
- Ferraty, F. and Vieu, P.: Nonparametric functional data analysis: theory and prac-tice. Springer, New York. (2006).Google Scholar
- Ramsay, J.O. and Silverman, B.W.: Functional data analysis (Second edition). Springer, New York. (2005).Google Scholar