hal-00564291, version 4
Sparsity considerations for dependent observations
(2011-02-08)
Résumé : The aim of this paper is to provide a comprehensive introduction for the study of L1-penalized estimators in the context of dependent observations. We define a general $\ell_{1}$-penalized estimator for solving problems of stochastic optimization. This estimator turns out to be the LASSO in the regression estimation setting. Powerful theoretical guarantees on the statistical performances of the LASSO were provided in recent papers, however, they usually only deal with the iid case. Here, we study our estimator under various dependence assumptions.
- 1 :
- CNRS : UMR7599 – Université Pierre et Marie Curie [UPMC] - Paris VI – Université Paris VII - Paris Diderot
- 2 :
- INSEE – École Nationale de la Statistique et de l'Administration Économique
- 3 :
- CNRS : UMR8088 – Université de Cergy Pontoise
- Domaine : Mathématiques/Statistiques
Statistiques/Théorie - Mots-clés : estimation in high dimension – weak dependence – sparsity – deviation of empirical mean – penalization – LASSO – regression estimation – density estimation.
- Versions disponibles : v1 (08-02-2011) v2 (17-02-2011) v3 (24-02-2011) v4 (06-07-2011) v5 (07-08-2011)
- hal-00564291, version 4
- http://hal.archives-ouvertes.fr/hal-00564291
- oai:hal.archives-ouvertes.fr:hal-00564291
- Contributeur :
- Soumis le : Mercredi 6 Juillet 2011, 16:24:02
- Dernière modification le : Mercredi 6 Juillet 2011, 17:05:51
Documents associés
Exporter