Sur la convergence de Dm-splines d’ajustement pour des données exactes ou bruitées
Departamento de Matemática Aplicada, CPS Calle María de Luna, 3 E-50015 Zaragoza — Spain | Laboratoire de Mathématiques Apliquées URA CNRS 1204, Université de Pau Avenue de l’Université F-64000 Pau — France |
Let be a bounded open subset of containing the data points, the Hausdorff distance from the set of data points to and the smoothing parameter. Assume that , , and in addition when the data are noisy, that the “white noise” hypothesis is satisfied. Then, for smoothing of exact or noisy data with -splines, convergence and error estimates results are obtained. These results improve previous ones given by F. Utreras or the authors.
1991 Mathematics Subject Classification: 41A15.