Managing uncertainty in complex stochastic models: design and emulation of a Rabies model

Alexis Boukouvalas, Dan Cornford, Alexander Singer

Research output: Contribution to conferencePaper

Abstract

In this paper we present a novel method for emulating a stochastic, or random output, computer model and show its application to a complex rabies model. The method is evaluated both in terms of accuracy and computational efficiency on synthetic data and the rabies model. We address the issue of experimental design and provide empirical evidence on the effectiveness of utilizing replicate model evaluations compared to a space-filling design. We employ the Mahalanobis error measure to validate the heteroscedastic Gaussian process based emulator predictions for both the mean and (co)variance. The emulator allows efficient screening to identify important model inputs and better understanding of the complex behaviour of the rabies model.
Original languageEnglish
Pages839-841
Number of pages3
Publication statusPublished - Jun 2009
Event6th St. Petersburg Workshop on Simulation - St Petersburg , Russian Federation
Duration: 28 Jun 20094 Jul 2009

Other

Other6th St. Petersburg Workshop on Simulation
CountryRussian Federation
CitySt Petersburg
Period28/06/094/07/09

Fingerprint

Stochastic models
Computational efficiency
Design of experiments
Uncertainty
Screening

Keywords

  • stochastic computer model
  • complex rabies mode
  • Mahalanobis error measure
  • heteroscedastic Gaussian process
  • emulator predictions

Cite this

Boukouvalas, A., Cornford, D., & Singer, A. (2009). Managing uncertainty in complex stochastic models: design and emulation of a Rabies model. 839-841. Paper presented at 6th St. Petersburg Workshop on Simulation, St Petersburg , Russian Federation.
Boukouvalas, Alexis ; Cornford, Dan ; Singer, Alexander. / Managing uncertainty in complex stochastic models: design and emulation of a Rabies model. Paper presented at 6th St. Petersburg Workshop on Simulation, St Petersburg , Russian Federation.3 p.
@conference{c28c670f9b0b4b5a8222137410137602,
title = "Managing uncertainty in complex stochastic models: design and emulation of a Rabies model",
abstract = "In this paper we present a novel method for emulating a stochastic, or random output, computer model and show its application to a complex rabies model. The method is evaluated both in terms of accuracy and computational efficiency on synthetic data and the rabies model. We address the issue of experimental design and provide empirical evidence on the effectiveness of utilizing replicate model evaluations compared to a space-filling design. We employ the Mahalanobis error measure to validate the heteroscedastic Gaussian process based emulator predictions for both the mean and (co)variance. The emulator allows efficient screening to identify important model inputs and better understanding of the complex behaviour of the rabies model.",
keywords = "stochastic computer model, complex rabies mode, Mahalanobis error measure, heteroscedastic Gaussian process, emulator predictions",
author = "Alexis Boukouvalas and Dan Cornford and Alexander Singer",
year = "2009",
month = "6",
language = "English",
pages = "839--841",
note = "6th St. Petersburg Workshop on Simulation ; Conference date: 28-06-2009 Through 04-07-2009",

}

Boukouvalas, A, Cornford, D & Singer, A 2009, 'Managing uncertainty in complex stochastic models: design and emulation of a Rabies model' Paper presented at 6th St. Petersburg Workshop on Simulation, St Petersburg , Russian Federation, 28/06/09 - 4/07/09, pp. 839-841.

Managing uncertainty in complex stochastic models: design and emulation of a Rabies model. / Boukouvalas, Alexis; Cornford, Dan; Singer, Alexander.

2009. 839-841 Paper presented at 6th St. Petersburg Workshop on Simulation, St Petersburg , Russian Federation.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Managing uncertainty in complex stochastic models: design and emulation of a Rabies model

AU - Boukouvalas, Alexis

AU - Cornford, Dan

AU - Singer, Alexander

PY - 2009/6

Y1 - 2009/6

N2 - In this paper we present a novel method for emulating a stochastic, or random output, computer model and show its application to a complex rabies model. The method is evaluated both in terms of accuracy and computational efficiency on synthetic data and the rabies model. We address the issue of experimental design and provide empirical evidence on the effectiveness of utilizing replicate model evaluations compared to a space-filling design. We employ the Mahalanobis error measure to validate the heteroscedastic Gaussian process based emulator predictions for both the mean and (co)variance. The emulator allows efficient screening to identify important model inputs and better understanding of the complex behaviour of the rabies model.

AB - In this paper we present a novel method for emulating a stochastic, or random output, computer model and show its application to a complex rabies model. The method is evaluated both in terms of accuracy and computational efficiency on synthetic data and the rabies model. We address the issue of experimental design and provide empirical evidence on the effectiveness of utilizing replicate model evaluations compared to a space-filling design. We employ the Mahalanobis error measure to validate the heteroscedastic Gaussian process based emulator predictions for both the mean and (co)variance. The emulator allows efficient screening to identify important model inputs and better understanding of the complex behaviour of the rabies model.

KW - stochastic computer model

KW - complex rabies mode

KW - Mahalanobis error measure

KW - heteroscedastic Gaussian process

KW - emulator predictions

M3 - Paper

SP - 839

EP - 841

ER -

Boukouvalas A, Cornford D, Singer A. Managing uncertainty in complex stochastic models: design and emulation of a Rabies model. 2009. Paper presented at 6th St. Petersburg Workshop on Simulation, St Petersburg , Russian Federation.