Efficient rare event simulation for heavy-tailed systems via cross entropy

详细信息    查看全文

• 作者：Jose Blanchet ; ^{jose.blanchet@columbia.edu} ; Yixi Shi
• 关键词：Rare-event simulation ; Importance sampling ; Cross entropy method
• 刊名：Operations Research Letters
• 出版年：2013
• 出版时间：May, 2013
• 年：2013
• 卷：41
• 期：3
• 页码：271-276
• 全文大小：316 K

文摘

The cross entropy method is an iterative technique that is used to obtain a low-variance importance sampling (IS) distribution from a given parametric family, which must satisfy two properties. First, subsequent iterations of the parameters must be easily computable and, second, the family should approximate the zero-variance IS distribution. We obtain parametric families for which these two properties are satisfied for a large class of heavy-tailed systems. Our estimators are shown to be strongly efficient in these settings.