Template-Type: ReDIF-Paper 1.0
Series: Tinbergen Institute Discussion Papers
Creation-Date: 2013-09-02
Number: 13-127/III
Author-Name: Zdravko Botev
Author-Workplace-Name: The University of New South Wales, Sydney, Australia
Author-Name: Ad Ridder
Author-Workplace-Name: VU University Amsterdam
Author-Name: Leonardo Rojas-Nandayapa
Author-Workplace-Name: The University of Queensland
Title: Semiparametric Cross Entropy for Rare-Event Simulation
Abstract: The Cross Entropy method is a well-known adaptive importance  sampling method for rare-event probability estimation, which requires estimating an optimal importance sampling density within a parametric class. In this article we estimate an optimal importance sampling density within a wider semiparametric class of distributions. We show that this semiparametric version of the Cross Entropy method frequently yields efficient estimators. We illustrate the excellent practical performance of the method with numerical experiments and show that for the problems we consider it typically outperforms alternative schemes by orders of magnitude.
Classification-JEL: C61, C63
Keywords: Light-Tailed; Regularly-Varying; Subexponential; Rare-Event Probability; Cross Entropy method, Markov Chain Monte Carlo
File-Url: https://papers.tinbergen.nl/13127.pdf
File-Format: application/pdf
File-Size: 234604 bytes
Handle: RePEc:tin:wpaper:20130127