New families of qb-optimal saturated two-level main effects screening designs

Pi Wen Tsai, Steven Gilmour

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, we study saturated two-level main effects designs which are commonly used for screening experiments. The QB criterion, which incorporates experimenters' prior beliefs about the probability of factors being active is used to compare designs. We show that under priors with more weight on models of small size, p-efficient designs should be recommended; when models with more parameters are of interest, A-optimal designs would be better. We identify new classes of saturated main effects designs between these two designs under different priors. The way in which the choice of designs depends on experimenters' prior beliefs is demonstrated for the cases when the number of runs N - 2 mod 4. A novel method of construction of QB-optimal designs using conference matrices is introduced. Complete families of optimal designs are given for N = 6; 10; 14; 18; 26; 30.

Original languageEnglish
Pages (from-to)605-617
Number of pages13
JournalStatistica Sinica
Volume26
Issue number2
DOIs
Publication statusPublished - 2016 Apr

Fingerprint

Screening Design
Main Effect
A-optimal Design
Screening Experiment
Family
Design
Screening
Model

Keywords

  • Conference matrix
  • Model uncertainty
  • Prior information
  • QB-criterion
  • Screening
  • Weighing design

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

New families of qb-optimal saturated two-level main effects screening designs. / Tsai, Pi Wen; Gilmour, Steven.

In: Statistica Sinica, Vol. 26, No. 2, 04.2016, p. 605-617.

Research output: Contribution to journalArticle

@article{6c7d85e9fb414acbb61a4d07f3a834aa,
title = "New families of qb-optimal saturated two-level main effects screening designs",
abstract = "In this paper, we study saturated two-level main effects designs which are commonly used for screening experiments. The QB criterion, which incorporates experimenters' prior beliefs about the probability of factors being active is used to compare designs. We show that under priors with more weight on models of small size, p-efficient designs should be recommended; when models with more parameters are of interest, A-optimal designs would be better. We identify new classes of saturated main effects designs between these two designs under different priors. The way in which the choice of designs depends on experimenters' prior beliefs is demonstrated for the cases when the number of runs N - 2 mod 4. A novel method of construction of QB-optimal designs using conference matrices is introduced. Complete families of optimal designs are given for N = 6; 10; 14; 18; 26; 30.",
keywords = "Conference matrix, Model uncertainty, Prior information, QB-criterion, Screening, Weighing design",
author = "Tsai, {Pi Wen} and Steven Gilmour",
year = "2016",
month = "4",
doi = "10.5705/ss.202015.0084",
language = "English",
volume = "26",
pages = "605--617",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "2",

}

TY - JOUR

T1 - New families of qb-optimal saturated two-level main effects screening designs

AU - Tsai, Pi Wen

AU - Gilmour, Steven

PY - 2016/4

Y1 - 2016/4

N2 - In this paper, we study saturated two-level main effects designs which are commonly used for screening experiments. The QB criterion, which incorporates experimenters' prior beliefs about the probability of factors being active is used to compare designs. We show that under priors with more weight on models of small size, p-efficient designs should be recommended; when models with more parameters are of interest, A-optimal designs would be better. We identify new classes of saturated main effects designs between these two designs under different priors. The way in which the choice of designs depends on experimenters' prior beliefs is demonstrated for the cases when the number of runs N - 2 mod 4. A novel method of construction of QB-optimal designs using conference matrices is introduced. Complete families of optimal designs are given for N = 6; 10; 14; 18; 26; 30.

AB - In this paper, we study saturated two-level main effects designs which are commonly used for screening experiments. The QB criterion, which incorporates experimenters' prior beliefs about the probability of factors being active is used to compare designs. We show that under priors with more weight on models of small size, p-efficient designs should be recommended; when models with more parameters are of interest, A-optimal designs would be better. We identify new classes of saturated main effects designs between these two designs under different priors. The way in which the choice of designs depends on experimenters' prior beliefs is demonstrated for the cases when the number of runs N - 2 mod 4. A novel method of construction of QB-optimal designs using conference matrices is introduced. Complete families of optimal designs are given for N = 6; 10; 14; 18; 26; 30.

KW - Conference matrix

KW - Model uncertainty

KW - Prior information

KW - QB-criterion

KW - Screening

KW - Weighing design

UR - http://www.scopus.com/inward/record.url?scp=85011357457&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011357457&partnerID=8YFLogxK

U2 - 10.5705/ss.202015.0084

DO - 10.5705/ss.202015.0084

M3 - Article

AN - SCOPUS:85011357457

VL - 26

SP - 605

EP - 617

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 2

ER -