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
Recommended Citation
Dong, Hui and Nakayama, Marvin K., "Quantile estimation using conditional Monte Carlo and Latin hypercube sampling" (2017). Faculty Publications. 9512.
https://digitalcommons.njit.edu/fac_pubs/9512
