Estimating orthant probabilities of high dimensional Gaussian vectors with an application to set estimation - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2016

Estimating orthant probabilities of high dimensional Gaussian vectors with an application to set estimation

Résumé

The computation of Gaussian orthant probabilities has been extensively studied for low dimensional vectors. Here we focus on the high dimensional case and we present a two step procedure relying on both deterministic and stochastic techniques. The proposed estimator relies indeed on splitting the probability into a low dimensional term and a remainder. While the low dimensional probability can be estimated by fast and accurate quadrature, the remainder requires Monte Carlo sampling. We show that an estimator obtained with this technique has higher efficiency than standard Monte Carlo methods. We further refine the estimation by using a novel asymmetric nested Monte Carlo algorithm for the remainder and we highlight cases where this approximation brings substantial efficiency gains. Finally this method is applied to derive conservative estimates of excursion sets of expensive to evaluate deterministic functions under a Gaussian random field prior without requiring a Markov assumption.
Fichier principal
Vignette du fichier
mainArticle.pdf (483.65 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01289126 , version 1 (16-03-2016)
hal-01289126 , version 2 (13-11-2016)
hal-01289126 , version 3 (30-11-2018)

Identifiants

Citer

Dario Azzimonti, David Ginsbourger. Estimating orthant probabilities of high dimensional Gaussian vectors with an application to set estimation. 2016. ⟨hal-01289126v1⟩

Collections

CNRS
294 Consultations
1681 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More