Better Confidence Intervals for Importance Sampling

It is well known that for highly skewed distributions the standard method of using the t statistic for the confidence interval of the mean does not give robust results. This is an important problem for importance sampling (IS) as its final distribution is often skewed due to a heavy tailed weight di...

Full description

Bibliographic Details
Main Authors: Sak, Halis, Hörmann, Wolfgang, Leydold, Josef
Format: Others
Language:en
Published: WU Vienna University of Economics and Business 2010
Subjects:
Online Access:http://epub.wu.ac.at/3017/1/techreport%2D106.pdf