Estimation of a cumulative distribution function under interval censoring ''case 1'' via warped wavelets
Résumé
The estimation of an unknown cumulative distribution function in the interval censoring ''case 1'' model from dependent sequences is considered. We construct a new adaptive estimator based on a warped wavelet basis and a hard thresholding rule. Under mild assumptions on the parameters of the model, considering the $\mathbb{L}_2$ risk and the weighted Besov balls, we prove that the estimator attains a sharp rate of convergence. We also investigate its practical performances thanks to simulation experiments.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...