A general criterion for factorial designs under model uncertainty

Pi Wen Tsai, Steven G. Gilmour

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Motivated by two industrial experiments in which rather extreme prior knowledge was used to choose the design, we show that the QB criterion, which aims to improve the estimation in as many models as possible by incorporating experimenters' prior knowledge along with an approximation to the As criterion, is more general and has a better statistical interpretation than many standard criteria. The generalization and application of the criterion to different types of designs are presented. The relationships between QB and other criteria for different situations are explored. It is shown that the E(s2) criterion is a special case of QB and several aberration-type criteria are limiting cases of our criterion, so that QB provides a bridge between alphabetic optimality and aberration. The two case studies illustrate the potential benefits of the QB criterion. R programs for calculating QB are available online as supplemental materials.

Original languageEnglish
Pages (from-to)231-242
Number of pages12
JournalTechnometrics
Volume52
Issue number2
DOIs
Publication statusPublished - 2010 May 1

Fingerprint

Factorial Design
Model Uncertainty
Aberrations
Aberration
Prior Knowledge
Experiments
Uncertainty
Optimality
Extremes
Limiting
Choose

Keywords

  • Aberration
  • Design optimality criterion
  • Generalized minimum aberration
  • Model robust
  • Projection efficiency
  • Supersaturated design

ASJC Scopus subject areas

  • Statistics and Probability
  • Modelling and Simulation
  • Applied Mathematics

Cite this

A general criterion for factorial designs under model uncertainty. / Tsai, Pi Wen; Gilmour, Steven G.

In: Technometrics, Vol. 52, No. 2, 01.05.2010, p. 231-242.

Research output: Contribution to journalArticle

Tsai, Pi Wen ; Gilmour, Steven G. / A general criterion for factorial designs under model uncertainty. In: Technometrics. 2010 ; Vol. 52, No. 2. pp. 231-242.
@article{24ca0446a3954ab0b5f89ba655e1a63d,
title = "A general criterion for factorial designs under model uncertainty",
abstract = "Motivated by two industrial experiments in which rather extreme prior knowledge was used to choose the design, we show that the QB criterion, which aims to improve the estimation in as many models as possible by incorporating experimenters' prior knowledge along with an approximation to the As criterion, is more general and has a better statistical interpretation than many standard criteria. The generalization and application of the criterion to different types of designs are presented. The relationships between QB and other criteria for different situations are explored. It is shown that the E(s2) criterion is a special case of QB and several aberration-type criteria are limiting cases of our criterion, so that QB provides a bridge between alphabetic optimality and aberration. The two case studies illustrate the potential benefits of the QB criterion. R programs for calculating QB are available online as supplemental materials.",
keywords = "Aberration, Design optimality criterion, Generalized minimum aberration, Model robust, Projection efficiency, Supersaturated design",
author = "Tsai, {Pi Wen} and Gilmour, {Steven G.}",
year = "2010",
month = "5",
day = "1",
doi = "10.1198/TECH.2010.08093",
language = "English",
volume = "52",
pages = "231--242",
journal = "Technometrics",
issn = "0040-1706",
publisher = "American Statistical Association",
number = "2",

}

TY - JOUR

T1 - A general criterion for factorial designs under model uncertainty

AU - Tsai, Pi Wen

AU - Gilmour, Steven G.

PY - 2010/5/1

Y1 - 2010/5/1

N2 - Motivated by two industrial experiments in which rather extreme prior knowledge was used to choose the design, we show that the QB criterion, which aims to improve the estimation in as many models as possible by incorporating experimenters' prior knowledge along with an approximation to the As criterion, is more general and has a better statistical interpretation than many standard criteria. The generalization and application of the criterion to different types of designs are presented. The relationships between QB and other criteria for different situations are explored. It is shown that the E(s2) criterion is a special case of QB and several aberration-type criteria are limiting cases of our criterion, so that QB provides a bridge between alphabetic optimality and aberration. The two case studies illustrate the potential benefits of the QB criterion. R programs for calculating QB are available online as supplemental materials.

AB - Motivated by two industrial experiments in which rather extreme prior knowledge was used to choose the design, we show that the QB criterion, which aims to improve the estimation in as many models as possible by incorporating experimenters' prior knowledge along with an approximation to the As criterion, is more general and has a better statistical interpretation than many standard criteria. The generalization and application of the criterion to different types of designs are presented. The relationships between QB and other criteria for different situations are explored. It is shown that the E(s2) criterion is a special case of QB and several aberration-type criteria are limiting cases of our criterion, so that QB provides a bridge between alphabetic optimality and aberration. The two case studies illustrate the potential benefits of the QB criterion. R programs for calculating QB are available online as supplemental materials.

KW - Aberration

KW - Design optimality criterion

KW - Generalized minimum aberration

KW - Model robust

KW - Projection efficiency

KW - Supersaturated design

UR - http://www.scopus.com/inward/record.url?scp=77955331368&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955331368&partnerID=8YFLogxK

U2 - 10.1198/TECH.2010.08093

DO - 10.1198/TECH.2010.08093

M3 - Article

AN - SCOPUS:77955331368

VL - 52

SP - 231

EP - 242

JO - Technometrics

JF - Technometrics

SN - 0040-1706

IS - 2

ER -