Bayesian D-optimal supersaturated designs

Bradley Jones; Dennis K.J. Lin; Christopher J. Nachtsheim

doi:10.1016/j.jspi.2007.05.021

Bayesian D-optimal supersaturated designs

Bradley Jones, Dennis K.J. Lin, Christopher J. Nachtsheim

Supply Chain and Operations

Research output: Contribution to journal › Article › peer-review

48 Scopus citations

Abstract

We introduce a new class of supersaturated designs using Bayesian D-optimality. The designs generated using this approach can have arbitrary sample sizes, can have any number of blocks of any size, and can incorporate categorical factors with more than two levels. In side by side diagnostic comparisons based on the E (s²) criterion for two-level experiments having even sample size, our designs either match or out-perform the best designs published to date. The generality of the method is illustrated with quality improvement experiment with 15 runs and 20 factors in 3 blocks.

Original language	English (US)
Pages (from-to)	86-92
Number of pages	7
Journal	Journal of Statistical Planning and Inference
Volume	138
Issue number	1
DOIs	https://doi.org/10.1016/j.jspi.2007.05.021
State	Published - Jan 1 2008

Bibliographical note

Copyright:
Copyright 2007 Elsevier B.V., All rights reserved.

Keywords

Blocking
Exchange algorithm
Model robust designs
Ridge regression
Screening designs

Access

10.1016/j.jspi.2007.05.021

OpenUrl availability

Full text

Cite this

@article{d6c57759377a47d3a49e978104ce76dd,

title = "Bayesian D-optimal supersaturated designs",

abstract = "We introduce a new class of supersaturated designs using Bayesian D-optimality. The designs generated using this approach can have arbitrary sample sizes, can have any number of blocks of any size, and can incorporate categorical factors with more than two levels. In side by side diagnostic comparisons based on the E (s2) criterion for two-level experiments having even sample size, our designs either match or out-perform the best designs published to date. The generality of the method is illustrated with quality improvement experiment with 15 runs and 20 factors in 3 blocks.",

keywords = "Blocking, Exchange algorithm, Model robust designs, Ridge regression, Screening designs",

author = "Bradley Jones and Lin, {Dennis K.J.} and Nachtsheim, {Christopher J.}",

year = "2008",

month = jan,

day = "1",

doi = "10.1016/j.jspi.2007.05.021",

language = "English (US)",

volume = "138",

pages = "86--92",

journal = "Journal of Statistical Planning and Inference",

issn = "0378-3758",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - Bayesian D-optimal supersaturated designs

AU - Jones, Bradley

AU - Lin, Dennis K.J.

AU - Nachtsheim, Christopher J.

PY - 2008/1/1

Y1 - 2008/1/1

N2 - We introduce a new class of supersaturated designs using Bayesian D-optimality. The designs generated using this approach can have arbitrary sample sizes, can have any number of blocks of any size, and can incorporate categorical factors with more than two levels. In side by side diagnostic comparisons based on the E (s2) criterion for two-level experiments having even sample size, our designs either match or out-perform the best designs published to date. The generality of the method is illustrated with quality improvement experiment with 15 runs and 20 factors in 3 blocks.

AB - We introduce a new class of supersaturated designs using Bayesian D-optimality. The designs generated using this approach can have arbitrary sample sizes, can have any number of blocks of any size, and can incorporate categorical factors with more than two levels. In side by side diagnostic comparisons based on the E (s2) criterion for two-level experiments having even sample size, our designs either match or out-perform the best designs published to date. The generality of the method is illustrated with quality improvement experiment with 15 runs and 20 factors in 3 blocks.

KW - Blocking

KW - Exchange algorithm

KW - Model robust designs

KW - Ridge regression

KW - Screening designs

UR - http://www.scopus.com/inward/record.url?scp=34548850048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548850048&partnerID=8YFLogxK

U2 - 10.1016/j.jspi.2007.05.021

DO - 10.1016/j.jspi.2007.05.021

M3 - Article

AN - SCOPUS:34548850048

SN - 0378-3758

VL - 138

SP - 86

EP - 92

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

IS - 1

ER -

Bayesian D-optimal supersaturated designs

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this