A non asymptotic variance theorem for unnormalized Feynman-Kac particle models

Frédéric Cérou 1 Pierre Del Moral 2, 3, 4 Arnaud Guyader 1, 5
1 ASPI - Applications of interacting particle systems to statistics
UR1 - Université de Rennes 1, Inria Rennes – Bretagne Atlantique , CNRS - Centre National de la Recherche Scientifique : UMR6074
2 A3 - Advanced analysis to code optimization
UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France
4 CQFD - Quality control and dynamic reliability
IMB - Institut de Mathématiques de Bordeaux, Inria Bordeaux - Sud-Ouest
Abstract : We present a non asymptotic theorem for interacting particle approximations of unnormalized Feynman-Kac models. We provide an original stochastic analysis based on Feynman-Kac semigroup techniques combined with recently developed coalescent tree-based functional representations of particle block distributions. We present some regularity conditions under which the $\LL_2$-relative error of these weighted particle measures grows linearly with respect to the time horizon yielding what seems to be the first results of this type for this class of unnormalized models. We also illustrate these results in the context of particle simulation of static Boltzmann-Gibbs measures and restricted distributions, with a special interest in rare event analysis.
Type de document :
Rapport
[Research Report] RR-6716, INRIA. 2008
Liste complète des métadonnées

Littérature citée [12 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00337392
Contributeur : Frederic Cerou <>
Soumis le : jeudi 6 novembre 2008 - 18:39:50
Dernière modification le : mardi 19 juin 2018 - 11:12:06
Document(s) archivé(s) le : mardi 9 octobre 2012 - 15:05:39

Fichiers

RR-6716.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00337392, version 1

Citation

Frédéric Cérou, Pierre Del Moral, Arnaud Guyader. A non asymptotic variance theorem for unnormalized Feynman-Kac particle models. [Research Report] RR-6716, INRIA. 2008. 〈inria-00337392〉

Partager

Métriques

Consultations de la notice

696

Téléchargements de fichiers

355