The University of Queensland Homepage
Takes you back to the UQ reSEARCHers Homepage You are at the UQ reSEARCHers site
  
      
   » Contacts
   » Profile
   » Organisation
   » Unit Memberships
   » Projects
   » Publications
   » Supervised Students
   » Projects Available
   » Print Profile

   » Search
   » Browse
   » Subscription
   » Home
   » About

  »  UQ Experts
  »  UQ locks in occupancy on Indooroopilly bushland property
May 22 2013
  »  Depression linked to almost doubled stroke risk in middle-aged women
May 17 2013
  »  UQ scientist wins top prize for battling disease
May 16 2013
  »  More Research News

Mathematics  
  

 Publication

2012  Chan, Joshua C.C. and Kroese, Dirk P. (2012) Improved cross-entropy method for estimation. Statistics and Computing, 22 5: 1-10.

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.

 Professor Dirk Kroese
eSpace Record:  
http://espace.library.uq.edu.au/view/UQ:256281

  
Keywords:  Cross-entropy, Variance minimization, Importance sampling, Kullback-Leibler divergence, Rare-event simulation, Likelihood ratio degeneracy, t copula
 
<< Back to Publication List

© 2006 The University of Queensland, Brisbane, Australia
ABN 63 942 912 684
CRICOS Provider No: 00025B
Authorised by: Deputy Vice Chancellor (Research)
Maintained by: rps@uq.edu.au