Quantile estimation using conditional Monte Carlo and Latin hypercube sampling

Document Type

Conference Proceeding

Publication Date

6-28-2017

Abstract

Quantiles are often employed to measure risk. We combine two variance-reduction techniques, conditional Monte Carlo and Latin hypercube sampling, to estimate a quantile. Compared to either method by itself, the combination can produce a quantile estimator with substantially smaller variance. In addition to devising a point estimator for the quantile when applying the combined approaches, we also describe how to construct confidence intervals for the quantile. Numerical results demonstrate the effectiveness of the methods.

Identifier

85044537574 (Scopus)

ISBN

[9781538634288]

Publication Title

Proceedings Winter Simulation Conference

External Full Text Location

https://doi.org/10.1109/WSC.2017.8247933

ISSN

08917736

First Page

1986

Last Page

1997

Grant

CMMI-1537322

Fund Ref

National Science Foundation

This document is currently not available here.

Share

COinS