Construction of orthogonal symmetric Latin hypercube designs

Lin Wang, Fasheng Sun, Dennis K.J. Lin, Min Qian Liu

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Latin hypercube designs (LHDs) have found wide application in computer experiments. It is known that orthogonal LHDs guarantee the orthogonality between all linear effects, and symmetric LHDs ensure the orthogonality between linear and second-order effects. In this paper, we propose a construction method for orthogonal symmetric LHDs. Most resulting LHDs can accommodate the maximum number of factors, thus can study many more factors than existing ones. Several methods for constructing nearly orthogonal symmetric LHDs are also provided. The constructed orthogonal and nearly orthogonal LHDs can be utilized to generate more nearly orthogonal symmetric LHDs. A detailed comparison with existing designs shows that the resulting designs have more flexible and economical run sizes, and many desirable design properties.

Original languageEnglish (US)
Pages (from-to)1503-1520
Number of pages18
JournalStatistica Sinica
Volume28
Issue number3
DOIs
StatePublished - Jul 1 2018

Fingerprint

Latin Hypercube Design
Symmetric Design
Orthogonal Design
Orthogonality
Computer Experiments
Linear Order

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Wang, Lin ; Sun, Fasheng ; Lin, Dennis K.J. ; Liu, Min Qian. / Construction of orthogonal symmetric Latin hypercube designs. In: Statistica Sinica. 2018 ; Vol. 28, No. 3. pp. 1503-1520.
@article{12138133f2174616b311a82bf3e52370,
title = "Construction of orthogonal symmetric Latin hypercube designs",
abstract = "Latin hypercube designs (LHDs) have found wide application in computer experiments. It is known that orthogonal LHDs guarantee the orthogonality between all linear effects, and symmetric LHDs ensure the orthogonality between linear and second-order effects. In this paper, we propose a construction method for orthogonal symmetric LHDs. Most resulting LHDs can accommodate the maximum number of factors, thus can study many more factors than existing ones. Several methods for constructing nearly orthogonal symmetric LHDs are also provided. The constructed orthogonal and nearly orthogonal LHDs can be utilized to generate more nearly orthogonal symmetric LHDs. A detailed comparison with existing designs shows that the resulting designs have more flexible and economical run sizes, and many desirable design properties.",
author = "Lin Wang and Fasheng Sun and Lin, {Dennis K.J.} and Liu, {Min Qian}",
year = "2018",
month = "7",
day = "1",
doi = "10.5705/ss.202017.0075",
language = "English (US)",
volume = "28",
pages = "1503--1520",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "3",

}

Construction of orthogonal symmetric Latin hypercube designs. / Wang, Lin; Sun, Fasheng; Lin, Dennis K.J.; Liu, Min Qian.

In: Statistica Sinica, Vol. 28, No. 3, 01.07.2018, p. 1503-1520.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Construction of orthogonal symmetric Latin hypercube designs

AU - Wang, Lin

AU - Sun, Fasheng

AU - Lin, Dennis K.J.

AU - Liu, Min Qian

PY - 2018/7/1

Y1 - 2018/7/1

N2 - Latin hypercube designs (LHDs) have found wide application in computer experiments. It is known that orthogonal LHDs guarantee the orthogonality between all linear effects, and symmetric LHDs ensure the orthogonality between linear and second-order effects. In this paper, we propose a construction method for orthogonal symmetric LHDs. Most resulting LHDs can accommodate the maximum number of factors, thus can study many more factors than existing ones. Several methods for constructing nearly orthogonal symmetric LHDs are also provided. The constructed orthogonal and nearly orthogonal LHDs can be utilized to generate more nearly orthogonal symmetric LHDs. A detailed comparison with existing designs shows that the resulting designs have more flexible and economical run sizes, and many desirable design properties.

AB - Latin hypercube designs (LHDs) have found wide application in computer experiments. It is known that orthogonal LHDs guarantee the orthogonality between all linear effects, and symmetric LHDs ensure the orthogonality between linear and second-order effects. In this paper, we propose a construction method for orthogonal symmetric LHDs. Most resulting LHDs can accommodate the maximum number of factors, thus can study many more factors than existing ones. Several methods for constructing nearly orthogonal symmetric LHDs are also provided. The constructed orthogonal and nearly orthogonal LHDs can be utilized to generate more nearly orthogonal symmetric LHDs. A detailed comparison with existing designs shows that the resulting designs have more flexible and economical run sizes, and many desirable design properties.

UR - http://www.scopus.com/inward/record.url?scp=85048703646&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048703646&partnerID=8YFLogxK

U2 - 10.5705/ss.202017.0075

DO - 10.5705/ss.202017.0075

M3 - Article

VL - 28

SP - 1503

EP - 1520

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 3

ER -