Improved cross-entropy method for estimation

Chan, Joshua C.C. and Kroese, Dirk P. (2012) Improved cross-entropy method for estimation. Statistics and Computing, 22 5: 1031-1040. doi:10.1007/s11222-011-9275-7


Author Chan, Joshua C.C.
Kroese, Dirk P.
Title Improved cross-entropy method for estimation
Journal name Statistics and Computing   Check publisher's open access policy
ISSN 0960-3174
1573-1375
Publication date 2012-01-01
Year available 2011
Sub-type Article (original research)
DOI 10.1007/s11222-011-9275-7
Open Access Status Not Open Access
Volume 22
Issue 5
Start page 1031
End page 1040
Total pages 10
Place of publication Secaucus, NJ , United States
Publisher Springer New York
Language eng
Abstract The cross-entropy (CE) method is an adaptive importance sampling procedure that has been successfully applied to a diverse range of complicated simulation problems. However, recent research has shown that in some high-dimensional settings, the likelihood ratio degeneracy problem becomes severe and the importance sampling estimator obtained from the CE algorithm becomes unreliable. We consider a variation of the CE method whose performance does not deteriorate as the dimension of the problem increases. We then illustrate the algorithm via a high-dimensional estimation problem in risk management.
Keyword Cross-entropy
Variance minimization
Importance sampling
Kullback-Leibler divergence
Rare-event simulation
Likelihood ratio degeneracy
t copula
Q-Index Code C1
Q-Index Status Confirmed Code
Institutional Status UQ
Additional Notes Published online 22 September 2011.

Document type: Journal Article
Sub-type: Article (original research)
Collections: School of Mathematics and Physics
Official 2012 Collection
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 13 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 19 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Mon, 17 Oct 2011, 18:23:56 EST by Prof Dirk P. Kroese on behalf of Mathematics