A new rejection sampling method for truncated multivariate Gaussian random variables restricted to convex sets

Abstract : Statistical researchers have shown increasing interest in generating truncated multivariate normal distributions. In this paper, we only assume that the acceptance region is convex and we focus on rejection sampling. We propose a new algorithm that outperforms crude rejection method for the simulation of truncated multivariate Gaussian random variables. The proposed algorithm is based on a generalization of Von Neumann's rejection technique which requires the determination of the mode of the truncated multivariate density function. We provide a theoretical upper bound for the ratio of the target probability density function over the proposal probability density function. The simulation results show that the method is especially efficient when the probability of the multivariate normal distribution of being inside the acceptance region is low.
Type de document :
Chapitre d'ouvrage
Ronald Cools and Dirk Nuyens. Monte Carlo and Quasi-Monte Carlo Methods , 163, pp.521-530, 2016, 〈https://link.springer.com/chapter/10.1007/978-3-319-33507-0_27〉
Liste complète des métadonnées

Littérature citée [27 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01063978
Contributeur : Hassan Maatouk <>
Soumis le : lundi 15 septembre 2014 - 11:21:41
Dernière modification le : jeudi 29 juin 2017 - 15:38:17
Document(s) archivé(s) le : mardi 16 décembre 2014 - 10:41:53

Fichier

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

Identifiants

  • HAL Id : hal-01063978, version 1

Collections

Citation

Hassan Maatouk, Xavier Bay. A new rejection sampling method for truncated multivariate Gaussian random variables restricted to convex sets. Ronald Cools and Dirk Nuyens. Monte Carlo and Quasi-Monte Carlo Methods , 163, pp.521-530, 2016, 〈https://link.springer.com/chapter/10.1007/978-3-319-33507-0_27〉. 〈hal-01063978〉

Partager

Métriques

Consultations de
la notice

452

Téléchargements du document

1276