On reduction of finite-sample variance by extended Latin hypercube sampling

DOI10.2307/3318470MaRDI QIDQ5933569zbMATH OpenOpenAlexFDO

Authors Nobuaki Hoshino, Akimichi Takemura

Publication date 26 July 2001

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://semanticscholar.org/paper/770d0c1d787e8106deb7b75f38c9ec105db4c202

zbMATH Keywords

Latin hypercube samplingMonte Carlo methodvariance reduction

Mathematics Subject Classification ID

Sampling theory, sample surveys (62D05) Analysis of variance and covariance (ANOVA) (62J10) Monte Carlo methods (65C05)

Review text

The problem of reducing the finite-sample variance is very important for the Monte Carlo technique, and this work gives new results in this direction. Theoretical investigations are supported by numerical simulations and this gives evidence concerning trustworthiness of theoretical conclusions. Nevertheless it is worth to mention that some citations are missing, I think the work by \textit{I. Sobol'} and \textit{A. Tutunnikov} [Math. Comput. Modelling 23, No. 8-9, 87-96 (1996; Zbl 0858.65024)] and other works of these authors are very close to the considered problem to be referred in this work.

Recommendations

• On Latin hypercube sampling
• A novel extension algorithm for optimized Latin hypercube sampling
• Some large deviations results for Latin hypercube sampling
• A fast general extension algorithm of Latin hypercube sampling
• scientific article; zbMATH DE number 410129
• Uniform Bound on Normal Approximation of Latin Hypercube Sampling

Cited in

(3)

• Some large deviations results for Latin hypercube sampling
• Latin hypercube sampling with dependence and applications in finance
• Ridge estimation of covariance matrix from data in two classes.

This page was built for publication: On reduction of finite-sample variance by extended Latin hypercube sampling

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5933569)