Nonparametric weighted estimators for biased data
Résumé
Several adaptive methods to estimate a density from biased data are pre-sented. Risk bounds for the estimators are provided and an empirical study is performed to compare various kernel and projection estimators associated with different adaptation methods, namely Lepski-type bandwidth selection in pointwise and global settings and model selection for projection estimators. A real data example taken from fluorescence lifetime measurements is also studied.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|
Loading...