2 Citations (Scopus)

Abstract

The use of definitive screening designs (DSDs) has been increasing since their introduction in 2011. These designs are used to screen factors and to make predictions. We assert that the choice of analysis method for these designs depends on the goal of the experiment, screening, or prediction. In this work, we present simulation results to address the explanatory (screening) use and the predictive use of DSDs. To address the predictive ability of DSDs, we use two 5-factor DSDs and simultaneously run central composite designs case studies on which we will compare several common analysis methods. Overall, we find that for screening purposes, the Dantzig selector using the Bayesian Information Criterion statistic is a good analysis choice; however, when the goal of analysis is prediction forward selection using the Bayesian Information Criterion statistic produces models with a lower mean squared prediction error. Copyright

Original languageEnglish (US)
JournalApplied Stochastic Models in Business and Industry
DOIs
StateAccepted/In press - Jan 1 2017

Fingerprint

Screening Design
Screening
Bayesian Information Criterion
Prediction
Statistic
Screening Experiment
Selector
Prediction Error
Mean Squared Error
Statistics
Composite
Design
Simulation
Composite materials

Keywords

  • Best subsets
  • Dantzig selector
  • Explanatory modeling
  • Forward selection
  • Predictive modeling
  • Test data

ASJC Scopus subject areas

  • Modeling and Simulation
  • Business, Management and Accounting(all)
  • Management Science and Operations Research

Cite this

Analysis of definitive screening designs : Screening vs prediction. / Weese, Maria L.; Ramsey, Philip J.; Montgomery, Douglas.

In: Applied Stochastic Models in Business and Industry, 01.01.2017.

Research output: Contribution to journalArticle

@article{9db80be377844a4282ccb73db1fdb539,
title = "Analysis of definitive screening designs: Screening vs prediction",
abstract = "The use of definitive screening designs (DSDs) has been increasing since their introduction in 2011. These designs are used to screen factors and to make predictions. We assert that the choice of analysis method for these designs depends on the goal of the experiment, screening, or prediction. In this work, we present simulation results to address the explanatory (screening) use and the predictive use of DSDs. To address the predictive ability of DSDs, we use two 5-factor DSDs and simultaneously run central composite designs case studies on which we will compare several common analysis methods. Overall, we find that for screening purposes, the Dantzig selector using the Bayesian Information Criterion statistic is a good analysis choice; however, when the goal of analysis is prediction forward selection using the Bayesian Information Criterion statistic produces models with a lower mean squared prediction error. Copyright",
keywords = "Best subsets, Dantzig selector, Explanatory modeling, Forward selection, Predictive modeling, Test data",
author = "Weese, {Maria L.} and Ramsey, {Philip J.} and Douglas Montgomery",
year = "2017",
month = "1",
day = "1",
doi = "10.1002/asmb.2297",
language = "English (US)",
journal = "Applied Stochastic Models in Business and Industry",
issn = "1524-1904",
publisher = "John Wiley and Sons Ltd",

}

TY - JOUR

T1 - Analysis of definitive screening designs

T2 - Screening vs prediction

AU - Weese, Maria L.

AU - Ramsey, Philip J.

AU - Montgomery, Douglas

PY - 2017/1/1

Y1 - 2017/1/1

N2 - The use of definitive screening designs (DSDs) has been increasing since their introduction in 2011. These designs are used to screen factors and to make predictions. We assert that the choice of analysis method for these designs depends on the goal of the experiment, screening, or prediction. In this work, we present simulation results to address the explanatory (screening) use and the predictive use of DSDs. To address the predictive ability of DSDs, we use two 5-factor DSDs and simultaneously run central composite designs case studies on which we will compare several common analysis methods. Overall, we find that for screening purposes, the Dantzig selector using the Bayesian Information Criterion statistic is a good analysis choice; however, when the goal of analysis is prediction forward selection using the Bayesian Information Criterion statistic produces models with a lower mean squared prediction error. Copyright

AB - The use of definitive screening designs (DSDs) has been increasing since their introduction in 2011. These designs are used to screen factors and to make predictions. We assert that the choice of analysis method for these designs depends on the goal of the experiment, screening, or prediction. In this work, we present simulation results to address the explanatory (screening) use and the predictive use of DSDs. To address the predictive ability of DSDs, we use two 5-factor DSDs and simultaneously run central composite designs case studies on which we will compare several common analysis methods. Overall, we find that for screening purposes, the Dantzig selector using the Bayesian Information Criterion statistic is a good analysis choice; however, when the goal of analysis is prediction forward selection using the Bayesian Information Criterion statistic produces models with a lower mean squared prediction error. Copyright

KW - Best subsets

KW - Dantzig selector

KW - Explanatory modeling

KW - Forward selection

KW - Predictive modeling

KW - Test data

UR - http://www.scopus.com/inward/record.url?scp=85037607044&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85037607044&partnerID=8YFLogxK

U2 - 10.1002/asmb.2297

DO - 10.1002/asmb.2297

M3 - Article

AN - SCOPUS:85037607044

JO - Applied Stochastic Models in Business and Industry

JF - Applied Stochastic Models in Business and Industry

SN - 1524-1904

ER -