 |
|
|
|
| HAL : hal-00407906, version 1 |
| arXiv : 0907.4728 |
| DOI : 10.1214/09-SS054 |
| Fiche détaillée |  Récupérer au format |
|
|
| Statistics Surveys 4 (2010) 40--79 |
|
|
|
|
| A survey of cross-validation procedures for model selection |
|
|
| Sylvain Arlot 1Alain Celisse 2 |
|
|
| ANR-09-JCJC-0027-01 Collaboration(s) |
|
|
| (2010) |
|
|
|
| Used to estimate the risk of an estimator or to perform model selection, cross-validation is a widespread strategy because of its simplicity and its apparent universality. Many results exist on the model selection performances of cross-validation procedures. This survey intends to relate these results to the most recent advances of model selection theory, with a particular emphasis on distinguishing empirical statements from rigorous theoretical results. As a conclusion, guidelines are provided for choosing the best cross-validation procedure according to the particular features of the problem in hand. |
|
|
|
|
|
|
|
|
|
|
|
| 1 : | Laboratoire d'informatique de l'école normale supérieure (LIENS) |
|
|
|
CNRS : UMR8548 – Ecole normale supérieure de Paris - ENS Paris |
|
|
| 2 : | Mathématiques et Informatique Appliquées (MIA) |
|
|
|
Institut national de la recherche agronomique (INRA) : UMR0518 – AgroParisTech |
|
|
|
|
|
|
|
|
|
| Domaine | : | Mathématiques/Statistiques Statistiques/Autres Statistiques/Théorie Statistiques/Applications Statistiques/Méthodologie |
|
|
| cross-validation – leave-one-out – model selection |
|
|
|
| Liste des fichiers attachés à ce document : | ||||||||||
|
||||||||||
|
|
|
| hal-00407906, version 1 | |
| http://hal.archives-ouvertes.fr/hal-00407906 | |
| oai:hal.archives-ouvertes.fr:hal-00407906 | |
| Contributeur : Alain Celisse | |
| Soumis le : Lundi 27 Juillet 2009, 19:24:11 | |
| Dernière modification le : Jeudi 28 Juin 2012, 10:17:03 | |
