Multivariate versus univariate Kriging metamodels for multi-response simulation models

Research output: Contribution to journalArticleScientificpeer-review

Abstract

To analyze the input/output behavior of simulation models with multiple responses, we may apply either univariate or multivariate Kriging (Gaussian process) metamodels. In multivariate Kriging we face a major problem: the covariance matrix of all responses should remain positive-definite; we therefore use the recently proposed “nonseparable dependence” model. To evaluate the performance of univariate and multivariate Kriging, we perform several Monte Carlo experiments that simulate Gaussian processes. These Monte Carlo results suggest that the simpler univariate Kriging gives smaller mean square error.
Original languageEnglish
Pages (from-to)573-582
JournalEuropean Journal of Operational Research
Volume236
Issue number2
Early online date10 Feb 2014
DOIs
Publication statusPublished - 16 Jul 2014

Fingerprint

Kriging
Metamodel
Univariate
Simulation Model
Covariance matrix
Gaussian Process
Mean square error
Multiple Responses
Monte Carlo Experiment
Nonseparable
Positive definite
Experiments
Simulation model
Evaluate
Output
Gaussian process
Model

Cite this

@article{15113f78327946a29ad214e780e9d88f,
title = "Multivariate versus univariate Kriging metamodels for multi-response simulation models",
abstract = "To analyze the input/output behavior of simulation models with multiple responses, we may apply either univariate or multivariate Kriging (Gaussian process) metamodels. In multivariate Kriging we face a major problem: the covariance matrix of all responses should remain positive-definite; we therefore use the recently proposed “nonseparable dependence” model. To evaluate the performance of univariate and multivariate Kriging, we perform several Monte Carlo experiments that simulate Gaussian processes. These Monte Carlo results suggest that the simpler univariate Kriging gives smaller mean square error.",
author = "Kleijnen, {Jack P.C.} and E. Mehdad",
year = "2014",
month = "7",
day = "16",
doi = "10.1016/j.ejor.2014.02.001",
language = "English",
volume = "236",
pages = "573--582",
journal = "European Journal of Operational Research",
issn = "0377-2217",
publisher = "Elsevier Science BV",
number = "2",

}

Multivariate versus univariate Kriging metamodels for multi-response simulation models. / Kleijnen, Jack P.C.; Mehdad, E.

In: European Journal of Operational Research, Vol. 236, No. 2, 16.07.2014, p. 573-582.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Multivariate versus univariate Kriging metamodels for multi-response simulation models

AU - Kleijnen, Jack P.C.

AU - Mehdad, E.

PY - 2014/7/16

Y1 - 2014/7/16

N2 - To analyze the input/output behavior of simulation models with multiple responses, we may apply either univariate or multivariate Kriging (Gaussian process) metamodels. In multivariate Kriging we face a major problem: the covariance matrix of all responses should remain positive-definite; we therefore use the recently proposed “nonseparable dependence” model. To evaluate the performance of univariate and multivariate Kriging, we perform several Monte Carlo experiments that simulate Gaussian processes. These Monte Carlo results suggest that the simpler univariate Kriging gives smaller mean square error.

AB - To analyze the input/output behavior of simulation models with multiple responses, we may apply either univariate or multivariate Kriging (Gaussian process) metamodels. In multivariate Kriging we face a major problem: the covariance matrix of all responses should remain positive-definite; we therefore use the recently proposed “nonseparable dependence” model. To evaluate the performance of univariate and multivariate Kriging, we perform several Monte Carlo experiments that simulate Gaussian processes. These Monte Carlo results suggest that the simpler univariate Kriging gives smaller mean square error.

U2 - 10.1016/j.ejor.2014.02.001

DO - 10.1016/j.ejor.2014.02.001

M3 - Article

VL - 236

SP - 573

EP - 582

JO - European Journal of Operational Research

JF - European Journal of Operational Research

SN - 0377-2217

IS - 2

ER -