Efficient importance sampling for Monte Carlo evaluation of exceedance probabilities

Simple item page
dc.creatorChan, Hock Peng
dc.creatorLai, Tze Leung
dc.date2007-03-30
dc.date.accessioned2026-07-07T07:55:06Z
dc.date.available2026-07-07T07:55:06Z
dc.descriptionLarge deviation theory has provided important clues for the choice of importance sampling measures for Monte Carlo evaluation of exceedance probabilities. However, Glasserman and Wang [Ann. Appl. Probab. 7 (1997) 731--746] have given examples in which importance sampling measures that are consistent with large deviations can perform much worse than direct Monte Carlo. We address this problem by using certain mixtures of exponentially twisted measures for importance sampling. Their asymptotic optimality is established by using a new class of likelihood ratio martingales and renewal theory.
dc.descriptionPublished at http://dx.doi.org/10.1214/105051606000000664 in the Annals of Applied Probability (http://www.imstat.org/aap/) by the Institute of Mathematical Statistics (http://www.imstat.org)
dc.identifierhttps://arxiv.org/abs/math/0703910
dc.identifierhttp://arxiv.org/abs/math/0703910
dc.identifierAnnals of Applied Probability 2007, Vol. 17, No. 2, 440-473
dc.identifierdoi:10.1214/105051606000000664
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/126853
dc.subjectProbability
dc.subject60F10, 65C05 (Primary) 60J05, 65C40 (Secondary)
dc.titleEfficient importance sampling for Monte Carlo evaluation of exceedance probabilities
dc.typetext

Files

Collections

arXiv