Generalized latin hypercube design for computer experiments

Dette, Holger; Pepelyshev, Andrey

Generalized latin hypercube design for computer experiments

dc.contributor.author	Dette, Holger	de
dc.contributor.author	Pepelyshev, Andrey	de
dc.date.accessioned	2009-10-29T10:24:19Z
dc.date.available	2009-10-29T10:24:19Z
dc.date.issued	2009-09-16	de
dc.description.abstract	Space filling designs, which satisfy a uniformity property, are widely used in computer experiments. In the present paper the performance of non-uniform experimental designs which locate more points in a neighborhood of the boundary of the design space is investigated. These designs are obtained by a quantile transformation of the one-dimensional projections of commonly used space filling designs. This transformation is motivated by logarithmic potential theory, which yields the arc-sine measure as equilibrium distribution. Alternative distance measures yield to Beta distributions, which put more weight in the interior of the design space. The methodology is illustrated for maximin Latin hypercube designs in several examples. In particular it is demonstrated that in many cases the new designs yield a smaller integrated mean square error for prediction. Moreover, the new designs yield to substantially better performance with respect to the entropy criterion.	en
dc.identifier.uri	http://hdl.handle.net/2003/26496
dc.identifier.uri	http://dx.doi.org/10.17877/DE290R-12659
dc.language.iso	en	de
dc.relation.ispartofseries	Discussion Paper / SFB 823; 24/2009	de
dc.subject	arc-sine distribution	en
dc.subject	design for computer experiments	en
dc.subject	Latin hypercube designs	en
dc.subject	logarithmic potential	en
dc.subject	space filling designs	en
dc.subject.ddc	310	de
dc.subject.ddc	330	de
dc.subject.ddc	620	de
dc.title	Generalized latin hypercube design for computer experiments	en
dc.type	Text	de
dc.type.publicationtype	report	de
dcterms.accessRights	open access

Files

Original bundle

Now showing 1 - 1 of 1

Name:: 024.pdf
Size:: 268.81 KB
Format:: Adobe Portable Document Format
Description:: DNB

Download

Collections

Sonderforschungsbereich (SFB) 823