Confidence Intervals for Randomized Quasi-Monte Carlo Estimators
Document Type
Conference Proceeding
Publication Date
1-1-2023
Abstract
Randomized Quasi-Monte Carlo (RQMC) methods provide unbiased estimators whose variance often converges at a faster rate than standard Monte Carlo as a function of the sample size. However, computing valid confidence intervals is challenging because the observations from a single randomization are dependent and the central limit theorem does not ordinarily apply. A natural solution is to replicate the RQMC process independently a small number of times to estimate the variance and use a standard confidence interval based on a normal or Student t distribution. We investigate the standard Student t approach and two bootstrap methods for getting nonparametric confidence intervals for the mean using a modest number of replicates. Our main conclusion is that intervals based on the Student t distribution are more reliable than even the bootstrap t method on the integration problems arising from RQMC.
Identifier
85185374259 (Scopus)
ISBN
[9798350369663]
Publication Title
Proceedings Winter Simulation Conference
External Full Text Location
https://doi.org/10.1109/WSC60868.2023.10408613
ISSN
08917736
First Page
445
Last Page
456
Grant
IIS-1837931
Fund Ref
National Science Foundation
Recommended Citation
L'Ecuyer, Pierre; Nakayama, Marvin K.; Owen, Art B.; and Tuffin, Bruno, "Confidence Intervals for Randomized Quasi-Monte Carlo Estimators" (2023). Faculty Publications. 2381.
https://digitalcommons.njit.edu/fac_pubs/2381