Cross Validation and Maximum Likelihood estimation of hyper-parameters of Gaussian processes with model misspecification

Abstract : The Maximum Likelihood (ML) and Cross Validation (CV) methods for estimating covariance hyper-parameters are compared, in the context of Kriging with a misspecified covariance structure. A two-step approach is used. First, the case of the estimation of a single variance hyper-parameter is addressed, for which the fixed correlation function is misspecified. A predictive variance based quality criterion is introduced and a closed-form expression of this criterion is derived. It is shown that when the correlation function is misspecified, the CV does better compared to ML, while ML is optimal when the model is well-specified. In the second step, the results of the first step are extended to the case when the hyper-parameters of the correlation function are also estimated from data.
Type de document :
Article dans une revue
Computational Statistics and Data Analysis, Elsevier, 2013, 66, pp.55-69
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-00905400
Contributeur : François Bachoc <>
Soumis le : lundi 18 novembre 2013 - 11:17:28
Dernière modification le : lundi 29 mai 2017 - 14:24:17
Document(s) archivé(s) le : mercredi 19 février 2014 - 05:45:18

Fichier

CV_ML_model_misspecification.p...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00905400, version 1

Collections

Citation

François Bachoc. Cross Validation and Maximum Likelihood estimation of hyper-parameters of Gaussian processes with model misspecification. Computational Statistics and Data Analysis, Elsevier, 2013, 66, pp.55-69. <hal-00905400>

Partager

Métriques

Consultations de
la notice

165

Téléchargements du document

246