Learning Out of Leaders

Abstract : This paper investigates the problem of selection and estimation in a high dimensional regression-type model. We propose a procedure with no optimization called LOL, for Learning Out of Leaders. LOL is an auto-driven algorithm with two thresholding steps. A first adaptive thresholding helps to select leaders among the initial regressors in such a way to reduce the dimensionality. Then a second thresholding follows the estimations and predictions performed by linear regression on the leaders. Theoretical results are proved. As an estimation procedure, LOL is optimal since the upper exponential bounds are achieved. Rates of convergence are provided and show that LOL is also consistent as a selection procedure. An extensive computational experiment is conducted to emphasize the practical good performances of LOL.
Type de document :
Pré-publication, Document de travail
2010
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00445690
Contributeur : Karine Tribouley <>
Soumis le : lundi 11 janvier 2010 - 10:22:13
Dernière modification le : jeudi 27 avril 2017 - 09:46:02
Document(s) archivé(s) le : vendredi 18 juin 2010 - 00:39:28

Fichiers

LOLarxiv.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00445690, version 1
  • ARXIV : 1001.1919

Collections

Citation

Gerard Kerkyacharian, Mathilde Mougeot, Dominique Picard, Karine Tribouley. Learning Out of Leaders. 2010. <hal-00445690>

Partager

Métriques

Consultations de
la notice

171

Téléchargements du document

81