Estimation of rare event probabilities using cross-entropy

Tito Homem-de-Mello, Reuven Y. Rubinstein

Research output: Contribution to journalConference articlepeer-review

32 Scopus citations

Abstract

This paper deals with estimation of probabilities of rare events in static simulation models using a fast adaptive two-stage procedure based on importance sampling and Kullback-Liebler's cross-entropy (CE). More specifically, at the first stage we estimate the optimal parameter vector in the importance sampling distribution using CE, and at the second stage we estimate the desired rare event probability using importance sampling (likelihood ratios). Some theoretical aspects of the proposed method, including its convergence, are established. The numerical results presented suggest that the method effectively estimates rare event probabilities.

Original languageEnglish
Pages (from-to)310-319
Number of pages10
JournalWinter Simulation Conference Proceedings
Volume1
StatePublished - 2002
Externally publishedYes
EventProceedings of the 2002 Winter Simulation Conference - San Diego, CA, United States
Duration: 8 Dec 200211 Dec 2002

Fingerprint

Dive into the research topics of 'Estimation of rare event probabilities using cross-entropy'. Together they form a unique fingerprint.

Cite this