Comparing computer experiments for the Gaussian process model using integrated prediction variance

Rachel T. Silvestrini, Douglas Montgomery, Bradley Jones

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

Space-filling designs are a common choice of experimental design strategy for computer experiments. This article compares space-filling design types based on their theoretical prediction variance properties with respect to the Gaussian process model. An analytical solution for calculating the integrated prediction variance (IV) of the Gaussian process model is given. Using the analytical calculation of IV as a response variable, this article presents a study of the effects of dimension; sample size; value of parameter vector, θ; and experimental design type using a factorial design and regression analysis.

Original languageEnglish (US)
Pages (from-to)164-174
Number of pages11
JournalQuality Engineering
Volume25
Issue number2
DOIs
StatePublished - 2013

Fingerprint

Design of experiments
Experiments
Regression analysis

Keywords

  • Computer simulation
  • Gaussian process models
  • Integrated variance
  • Space-filling designs

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Industrial and Manufacturing Engineering

Cite this

Comparing computer experiments for the Gaussian process model using integrated prediction variance. / Silvestrini, Rachel T.; Montgomery, Douglas; Jones, Bradley.

In: Quality Engineering, Vol. 25, No. 2, 2013, p. 164-174.

Research output: Contribution to journalArticle

@article{b633c31cf06a4da681dc5425ff6254a5,
title = "Comparing computer experiments for the Gaussian process model using integrated prediction variance",
abstract = "Space-filling designs are a common choice of experimental design strategy for computer experiments. This article compares space-filling design types based on their theoretical prediction variance properties with respect to the Gaussian process model. An analytical solution for calculating the integrated prediction variance (IV) of the Gaussian process model is given. Using the analytical calculation of IV as a response variable, this article presents a study of the effects of dimension; sample size; value of parameter vector, θ; and experimental design type using a factorial design and regression analysis.",
keywords = "Computer simulation, Gaussian process models, Integrated variance, Space-filling designs",
author = "Silvestrini, {Rachel T.} and Douglas Montgomery and Bradley Jones",
year = "2013",
doi = "10.1080/08982112.2012.758284",
language = "English (US)",
volume = "25",
pages = "164--174",
journal = "Quality Engineering",
issn = "0898-2112",
publisher = "Taylor and Francis Ltd.",
number = "2",

}

TY - JOUR

T1 - Comparing computer experiments for the Gaussian process model using integrated prediction variance

AU - Silvestrini, Rachel T.

AU - Montgomery, Douglas

AU - Jones, Bradley

PY - 2013

Y1 - 2013

N2 - Space-filling designs are a common choice of experimental design strategy for computer experiments. This article compares space-filling design types based on their theoretical prediction variance properties with respect to the Gaussian process model. An analytical solution for calculating the integrated prediction variance (IV) of the Gaussian process model is given. Using the analytical calculation of IV as a response variable, this article presents a study of the effects of dimension; sample size; value of parameter vector, θ; and experimental design type using a factorial design and regression analysis.

AB - Space-filling designs are a common choice of experimental design strategy for computer experiments. This article compares space-filling design types based on their theoretical prediction variance properties with respect to the Gaussian process model. An analytical solution for calculating the integrated prediction variance (IV) of the Gaussian process model is given. Using the analytical calculation of IV as a response variable, this article presents a study of the effects of dimension; sample size; value of parameter vector, θ; and experimental design type using a factorial design and regression analysis.

KW - Computer simulation

KW - Gaussian process models

KW - Integrated variance

KW - Space-filling designs

UR - http://www.scopus.com/inward/record.url?scp=84878173992&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878173992&partnerID=8YFLogxK

U2 - 10.1080/08982112.2012.758284

DO - 10.1080/08982112.2012.758284

M3 - Article

AN - SCOPUS:84878173992

VL - 25

SP - 164

EP - 174

JO - Quality Engineering

JF - Quality Engineering

SN - 0898-2112

IS - 2

ER -