Strongly Efficient Estimators for Light-tailed Sums (with P. Glynn)

Summary: When we started working on this problem we first concentrated on Gaussian random walks and used the Bahadur approximation to mimic the zero-variance importance sampling distribution, better results are obtained in the MCQMC 2008 version of this paper. The links provided above (just click on “Strongly Efficient Estimators for Light-tailed Sums” take you to MCQMC 2008 citation and the paper itself).