Construction, Properties, and Analysis of Group-Orthogonal Supersaturated Designs

Bradley Jones, Ryan Lekivetz, Dibyen Majumdar, Christopher J. Nachtsheim, Jonathan W. Stallrich

Research output: Contribution to journalArticle

Abstract

In this article, we propose a new method for constructing supersaturated designs that is based on the Kronecker product of two carefully chosen matrices. The construction method leads to a partitioning of the factors of the design such that the factors within a group are correlated to the others within the same group, but are orthogonal to any factor in any other group. We refer to the resulting designs as group-orthogonal supersaturated designs. We leverage this group structure to obtain an unbiased estimate of the error variance, and to develop an effective, design-based model selection procedure. Simulation results show that the use of these designs, in conjunction with our model selection procedure enables the identification of larger numbers of active main effects than have previously been reported for supersaturated designs. The designs can also be used in group screening; however, unlike previous group-screening procedures, with our designs, main effects in a group are not confounded. Supplementary materials for this article are available online.

Original languageEnglish (US)
JournalTechnometrics
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Supersaturated Design
Orthogonal Design
Main Effect
Selection Procedures
Model Selection
Screening
Kronecker Product
Leverage
Design
Partitioning
Identification (control systems)

Keywords

  • E(s2)-optimality
  • Group screening designs
  • Hadamard matrices
  • Model selection
  • UE-optimality

Cite this

Construction, Properties, and Analysis of Group-Orthogonal Supersaturated Designs. / Jones, Bradley; Lekivetz, Ryan; Majumdar, Dibyen; Nachtsheim, Christopher J.; Stallrich, Jonathan W.

In: Technometrics, 01.01.2019.

Research output: Contribution to journalArticle

Jones, Bradley ; Lekivetz, Ryan ; Majumdar, Dibyen ; Nachtsheim, Christopher J. ; Stallrich, Jonathan W. / Construction, Properties, and Analysis of Group-Orthogonal Supersaturated Designs. In: Technometrics. 2019.
@article{481301fd44084d5d9414ff0cbc724c78,
title = "Construction, Properties, and Analysis of Group-Orthogonal Supersaturated Designs",
abstract = "In this article, we propose a new method for constructing supersaturated designs that is based on the Kronecker product of two carefully chosen matrices. The construction method leads to a partitioning of the factors of the design such that the factors within a group are correlated to the others within the same group, but are orthogonal to any factor in any other group. We refer to the resulting designs as group-orthogonal supersaturated designs. We leverage this group structure to obtain an unbiased estimate of the error variance, and to develop an effective, design-based model selection procedure. Simulation results show that the use of these designs, in conjunction with our model selection procedure enables the identification of larger numbers of active main effects than have previously been reported for supersaturated designs. The designs can also be used in group screening; however, unlike previous group-screening procedures, with our designs, main effects in a group are not confounded. Supplementary materials for this article are available online.",
keywords = "E(s2)-optimality, Group screening designs, Hadamard matrices, Model selection, UE-optimality",
author = "Bradley Jones and Ryan Lekivetz and Dibyen Majumdar and Nachtsheim, {Christopher J.} and Stallrich, {Jonathan W.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/00401706.2019.1654926",
language = "English (US)",
journal = "Technometrics",
issn = "0040-1706",
publisher = "American Statistical Association",

}

TY - JOUR

T1 - Construction, Properties, and Analysis of Group-Orthogonal Supersaturated Designs

AU - Jones, Bradley

AU - Lekivetz, Ryan

AU - Majumdar, Dibyen

AU - Nachtsheim, Christopher J.

AU - Stallrich, Jonathan W.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this article, we propose a new method for constructing supersaturated designs that is based on the Kronecker product of two carefully chosen matrices. The construction method leads to a partitioning of the factors of the design such that the factors within a group are correlated to the others within the same group, but are orthogonal to any factor in any other group. We refer to the resulting designs as group-orthogonal supersaturated designs. We leverage this group structure to obtain an unbiased estimate of the error variance, and to develop an effective, design-based model selection procedure. Simulation results show that the use of these designs, in conjunction with our model selection procedure enables the identification of larger numbers of active main effects than have previously been reported for supersaturated designs. The designs can also be used in group screening; however, unlike previous group-screening procedures, with our designs, main effects in a group are not confounded. Supplementary materials for this article are available online.

AB - In this article, we propose a new method for constructing supersaturated designs that is based on the Kronecker product of two carefully chosen matrices. The construction method leads to a partitioning of the factors of the design such that the factors within a group are correlated to the others within the same group, but are orthogonal to any factor in any other group. We refer to the resulting designs as group-orthogonal supersaturated designs. We leverage this group structure to obtain an unbiased estimate of the error variance, and to develop an effective, design-based model selection procedure. Simulation results show that the use of these designs, in conjunction with our model selection procedure enables the identification of larger numbers of active main effects than have previously been reported for supersaturated designs. The designs can also be used in group screening; however, unlike previous group-screening procedures, with our designs, main effects in a group are not confounded. Supplementary materials for this article are available online.

KW - E(s2)-optimality

KW - Group screening designs

KW - Hadamard matrices

KW - Model selection

KW - UE-optimality

UR - http://www.scopus.com/inward/record.url?scp=85073934534&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073934534&partnerID=8YFLogxK

U2 - 10.1080/00401706.2019.1654926

DO - 10.1080/00401706.2019.1654926

M3 - Article

AN - SCOPUS:85073934534

JO - Technometrics

JF - Technometrics

SN - 0040-1706

ER -