2018
DOI: 10.1214/17-aos1674
|View full text |Cite
|
Sign up to set email alerts
|

Optimal maximin $L_{1}$-distance Latin hypercube designs based on good lattice point designs

Abstract: Maximin distance Latin hypercube designs are commonly used for computer experiments, but the construction of such designs is challenging. We construct a series of maximin Latin hypercube designs via Williams transformations of good lattice point designs. Some constructed designs are optimal under the maximin L 1 -distance criterion, while others are asymptotically optimal. Moreover, these designs are also shown to have small pairwise correlations between columns.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
33
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
7

Relationship

3
4

Authors

Journals

citations
Cited by 47 publications
(33 citation statements)
references
References 30 publications
0
33
0
Order By: Relevance
“…Orthogonality [Owen (1994), Tang (1998), Ye (1998)] and maximin distance criterion [Johnson, Moore and Ylvisaker (1990), Zhou and Xu (2014)] are two most commonly used design criteria. A number of authors have constructed orthogonal and maximin LHDs; see Lin and Tang (2015) for an excellent review, as well as Sun and Tang (2017), Xu (2017, 2018) and Wang, Xiao and Xu (2018) for some recent works. However, as Joseph and Hung (2008) and Xiao and Xu (2018) pointed out, orthogonal LHDs often do not have maximin distance and maximin distance designs are often not orthogonal, except for some specific cases [Wang, Yang and Xu (2018)].…”
Section: Introductionmentioning
confidence: 99%
“…Orthogonality [Owen (1994), Tang (1998), Ye (1998)] and maximin distance criterion [Johnson, Moore and Ylvisaker (1990), Zhou and Xu (2014)] are two most commonly used design criteria. A number of authors have constructed orthogonal and maximin LHDs; see Lin and Tang (2015) for an excellent review, as well as Sun and Tang (2017), Xu (2017, 2018) and Wang, Xiao and Xu (2018) for some recent works. However, as Joseph and Hung (2008) and Xiao and Xu (2018) pointed out, orthogonal LHDs often do not have maximin distance and maximin distance designs are often not orthogonal, except for some specific cases [Wang, Yang and Xu (2018)].…”
Section: Introductionmentioning
confidence: 99%
“…Due to the complexity of underlying biological systems, a systematic quantification of effects for multiple drugs is challenging, and thus, various models should be explored for such experiments. In such situations, space‐filling designs are ideal due to their robustness . Maximin distance designs are ideal for kriging models, as any unobserved point will not be too far from observed design points and thus the prediction error will not be too big.…”
Section: Conclusion and Discussionmentioning
confidence: 99%
“…In such situations, space-filling designs are ideal due to their robustness. [20][21][22][23][24] Maximin distance designs are ideal for kriging models, as any unobserved point will not be too far from observed design points and thus the prediction error will not be too big. An interesting topic for the future research is how space-filling designs, especially maximin distance designs, perform under kriging models in drug combination studies.…”
Section: Conclusion and Discussionmentioning
confidence: 99%
“…Note that linearly permuted regular designs can be still considered as regular because they are just cosets of regular designs and share the same defining relationship. We consider a nonlinear level permutation via the Williams transformation, which was first used by Williams (1949) to construct balanced Latin square designs, followed by Butler (2001) and Wang et al (2018b) to construct orthogonal or maximin LHDs. Our purpose is different from theirs.…”
Section: Introductionmentioning
confidence: 99%