Analysis of definitive screening designs: Screening vs prediction

Maria L. Weese, Philip J. Ramsey, Douglas Montgomery

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


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.

Original languageEnglish (US)
Pages (from-to)244-255
Number of pages12
JournalApplied Stochastic Models in Business and Industry
Issue number2
StatePublished - Mar 1 2018


  • Dantzig selector
  • best subsets
  • explanatory modeling
  • forward selection
  • predictive modeling
  • test data

ASJC Scopus subject areas

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


Dive into the research topics of 'Analysis of definitive screening designs: Screening vs prediction'. Together they form a unique fingerprint.

Cite this