Some large deviations results for Latin Hypercube sampling

Shane S. Drew, Tito Homem-de-Mello

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

11 Scopus citations


Large deviations theory is a well-studied area which has shown to have numerous applications. The typical results, however, assume that the underlying random variables are either i.i.d. or exhibit some form of Markovian dependence. Our interest in this paper is to study the validity of large deviations results in the context of estimators built with Latin Hypercube sampling, a well-known sampling technique for variance reduction. We show that a large deviation principle holds for Latin Hypercube sampling for functions in one dimension and for separable multi-dimensional functions. Moreover, the upper bound of the probability of a large deviation in these cases is no higher under Latin Hypercube sampling than it is under Monte Carlo sampling. We extend the latter property to functions that preserve negative dependence (such as functions that are monotone in each argument). Numerical experiments illustrate the theoretical results presented in the paper.

Original languageEnglish
Title of host publicationProceedings of the 2005 Winter Simulation Conference
Number of pages9
StatePublished - 2005
Externally publishedYes
Event2005 Winter Simulation Conference - Orlando, FL, United States
Duration: 4 Dec 20057 Dec 2005

Publication series

NameProceedings - Winter Simulation Conference
ISSN (Print)0891-7736


Conference2005 Winter Simulation Conference
Country/TerritoryUnited States
CityOrlando, FL


Dive into the research topics of 'Some large deviations results for Latin Hypercube sampling'. Together they form a unique fingerprint.

Cite this