Augmenting supersaturated designs with Bayesian D-optimality

Alex J. Gutman; Edward D. White; Dennis K.J. Lin; Raymond R. Hill

doi:10.1016/j.csda.2013.09.009

Augmenting supersaturated designs with Bayesian D-optimality

Alex J. Gutman, Edward D. White, Dennis K.J. Lin, Raymond R. Hill

Statistics

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

5 Scopus citations

Abstract

A methodology is developed to add runs to existing supersaturated designs. The technique uses information from the analysis of the initial experiment to choose the best possible follow-up runs. After analysis of the initial data, factors are classified into one of three groups: primary, secondary, and potential. Runs are added to maximize a Bayesian D-optimality criterion to increase the information gained about those factors. Simulation results show the method can outperform existing supersaturated design augmentation strategies that add runs without analyzing the initial response variables.

Original language	English (US)
Pages (from-to)	1147-1158
Number of pages	12
Journal	Computational Statistics and Data Analysis
Volume	71
DOIs	https://doi.org/10.1016/j.csda.2013.09.009
State	Published - Mar 2014

All Science Journal Classification (ASJC) codes

Statistics and Probability
Computational Mathematics
Computational Theory and Mathematics
Applied Mathematics

Access to Document

10.1016/j.csda.2013.09.009

Cite this

@article{67fbca83ada94bbbbec13cbcbc320823,

title = "Augmenting supersaturated designs with Bayesian D-optimality",

abstract = "A methodology is developed to add runs to existing supersaturated designs. The technique uses information from the analysis of the initial experiment to choose the best possible follow-up runs. After analysis of the initial data, factors are classified into one of three groups: primary, secondary, and potential. Runs are added to maximize a Bayesian D-optimality criterion to increase the information gained about those factors. Simulation results show the method can outperform existing supersaturated design augmentation strategies that add runs without analyzing the initial response variables.",

author = "Gutman, {Alex J.} and White, {Edward D.} and Lin, {Dennis K.J.} and Hill, {Raymond R.}",

note = "Funding Information: The authors would like to thank the Associate Editor and referees for many helpful suggestions. This research was supported by the Office of the Secretary of Defense, Director of Operational Test and Evaluation (OSD DOT&E) and the Test Resource Management Center (TRMC) under the Science of Test research program . Special thanks to both Dr. Catherine Warner, OSD DOT&E, and Mr. George Rumford, TRMC, for sponsoring the research program. ",

year = "2014",

month = mar,

doi = "10.1016/j.csda.2013.09.009",

language = "English (US)",

volume = "71",

pages = "1147--1158",

journal = "Computational Statistics and Data Analysis",

issn = "0167-9473",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Augmenting supersaturated designs with Bayesian D-optimality

AU - Gutman, Alex J.

AU - White, Edward D.

AU - Lin, Dennis K.J.

AU - Hill, Raymond R.

N1 - Funding Information: The authors would like to thank the Associate Editor and referees for many helpful suggestions. This research was supported by the Office of the Secretary of Defense, Director of Operational Test and Evaluation (OSD DOT&E) and the Test Resource Management Center (TRMC) under the Science of Test research program . Special thanks to both Dr. Catherine Warner, OSD DOT&E, and Mr. George Rumford, TRMC, for sponsoring the research program.

PY - 2014/3

Y1 - 2014/3

N2 - A methodology is developed to add runs to existing supersaturated designs. The technique uses information from the analysis of the initial experiment to choose the best possible follow-up runs. After analysis of the initial data, factors are classified into one of three groups: primary, secondary, and potential. Runs are added to maximize a Bayesian D-optimality criterion to increase the information gained about those factors. Simulation results show the method can outperform existing supersaturated design augmentation strategies that add runs without analyzing the initial response variables.

AB - A methodology is developed to add runs to existing supersaturated designs. The technique uses information from the analysis of the initial experiment to choose the best possible follow-up runs. After analysis of the initial data, factors are classified into one of three groups: primary, secondary, and potential. Runs are added to maximize a Bayesian D-optimality criterion to increase the information gained about those factors. Simulation results show the method can outperform existing supersaturated design augmentation strategies that add runs without analyzing the initial response variables.

UR - http://www.scopus.com/inward/record.url?scp=84889079891&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84889079891&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2013.09.009

DO - 10.1016/j.csda.2013.09.009

M3 - Article

AN - SCOPUS:84889079891

SN - 0167-9473

VL - 71

SP - 1147

EP - 1158

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

ER -

Augmenting supersaturated designs with Bayesian D-optimality

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this