L2-Boosting for sensitivity analysis with dependent inputs

Abstract : This paper is dedicated to the study of an estimator of the generalized Hoeffding decomposition. We build such an estimator using an empirical Gram-Schmidt approach and derive a consistency rate in a large dimensional setting. We then apply a greedy algorithm with these previous estimators to a sensitivity analysis. We also establish the consistency of this L2-boosting under sparsity assumptions of the signal to be analyzed. The paper concludes with numerical experiments, that demonstrate the low computational cost of our method, as well as its efficiency on the standard benchmark of sensitivity analysis.
Type de document :
Article dans une revue
Statistica Sinica, Taipei : Institute of Statistical Science, Academia Sinica, 2015, 25 (4), pp.1477-1502. 〈10.5705/ss.2013.310〉
Liste complète des métadonnées

Littérature citée [36 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01068729
Contributeur : Clémentine Prieur <>
Soumis le : vendredi 26 septembre 2014 - 16:27:23
Dernière modification le : mercredi 23 mai 2018 - 17:58:04
Document(s) archivé(s) le : samedi 27 décembre 2014 - 10:50:40

Fichier

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

Identifiants

Citation

Magali Champion, Gaëlle Chastaing, Sebastien Gadat, Clémentine Prieur. L2-Boosting for sensitivity analysis with dependent inputs. Statistica Sinica, Taipei : Institute of Statistical Science, Academia Sinica, 2015, 25 (4), pp.1477-1502. 〈10.5705/ss.2013.310〉. 〈hal-01068729〉

Partager

Métriques

Consultations de la notice

1162

Téléchargements de fichiers

279