Bayesian lightweight emulators for multivariate computer models
Bayesian lightweight emulators for multivariate computer models
Statistical emulators for the outputs of complex computer codes (simulators) are typically constructed using nonparametric regression methods, such as Gaussian Process (GP) regression. For many simulators, emulators based on parametric models may provide adequate descriptions whilst enabling straightforward and computationally inexpensive fitting, inference and prediction. We place such so called “lightweight” emulators into the same Bayesian framework as the more usual nonparametric emulators, and provide methodology for their application to two novel examples with multivariate output: an emergency-relief simulator and a low-level atmospheric dispersion simulator. For the former, the inputs to the simulator are both continuous and categorical, and a comparison is made to GP emulators; for the latter, the output is zeroinflated and an appropriate emulator is developed from a Tobit model. In each case, sensitivity analyses are performed to identify the inputs to the simulator that have a substantive impact on the response, using both traditional methods and Bayesian model selection.
bayesian linear regression, gaussian process, markov chain monte carlo model composition, tobit model, zero-inflated response
Southampton Statistical Sciences Research Institute, University of Southampton
Overstall, Antony M.
74c8d426-044e-4b3b-a5b3-23c9250c1a96
Woods, David C.
ae21f7e2-29d9-4f55-98a2-639c5e44c79c
14 January 2011
Overstall, Antony M.
74c8d426-044e-4b3b-a5b3-23c9250c1a96
Woods, David C.
ae21f7e2-29d9-4f55-98a2-639c5e44c79c
Overstall, Antony M. and Woods, David C.
(2011)
Bayesian lightweight emulators for multivariate computer models
(S3RI Methodology Working Papers, M11/01)
Southampton, GB.
Southampton Statistical Sciences Research Institute, University of Southampton
27pp.
Record type:
Monograph
(Working Paper)
Abstract
Statistical emulators for the outputs of complex computer codes (simulators) are typically constructed using nonparametric regression methods, such as Gaussian Process (GP) regression. For many simulators, emulators based on parametric models may provide adequate descriptions whilst enabling straightforward and computationally inexpensive fitting, inference and prediction. We place such so called “lightweight” emulators into the same Bayesian framework as the more usual nonparametric emulators, and provide methodology for their application to two novel examples with multivariate output: an emergency-relief simulator and a low-level atmospheric dispersion simulator. For the former, the inputs to the simulator are both continuous and categorical, and a comparison is made to GP emulators; for the latter, the output is zeroinflated and an appropriate emulator is developed from a Tobit model. In each case, sensitivity analyses are performed to identify the inputs to the simulator that have a substantive impact on the response, using both traditional methods and Bayesian model selection.
Text
s3ri-workingpaper-M11-01.pdf
- Other
More information
Published date: 14 January 2011
Keywords:
bayesian linear regression, gaussian process, markov chain monte carlo model composition, tobit model, zero-inflated response
Identifiers
Local EPrints ID: 171217
URI: http://eprints.soton.ac.uk/id/eprint/171217
PURE UUID: 2de196c3-6851-471c-ab33-62d5e553373b
Catalogue record
Date deposited: 14 Jan 2011 15:31
Last modified: 14 Mar 2024 02:44
Export record
Contributors
Author:
Antony M. Overstall
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
