Multiple predictor smoothing methods for sensitivity analysis

Curtis Storlie, Jon C. Helton

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

The use of multiple predictor smoothing methods in sampling-based sensitivity analyses of complex models is investigated. Specifically, sensitivity analysis procedures based on smoothing methods employing the stepwise application of the following nonparametric regression techniques are described: (i) locally weighted regression (LOESS), (ii) additive models (GAMs), (iii) projection pursuit regression (PP_REG), and (iv) recursive partitioning regression (RP_REG). The indicated procedures are illustrated with both simple test problems and results from a performance assessment for a radioactive waste disposal facility (i.e., the Waste Isolation Pilot Plant). As shown by the example illustrations, the use of smoothing procedures based on nonparametric regression techniques can yield more informative sensitivity analysis results than can be obtained with more traditional sensitivity analysis procedures based on linear regression, rank regression or response surface regression when nonlinear relationships between model inputs and model predictions are present.

Original languageEnglish (US)
Title of host publicationProceedings of the 2005 Winter Simulation Conference
Pages231-239
Number of pages9
Volume2005
DOIs
StatePublished - 2005
Externally publishedYes
Event2005 Winter Simulation Conference - Orlando, FL, United States
Duration: Dec 4 2005Dec 7 2005

Other

Other2005 Winter Simulation Conference
CountryUnited States
CityOrlando, FL
Period12/4/0512/7/05

Fingerprint

Sensitivity analysis
Radioactive waste disposal
Pilot plants
Linear regression
Sampling

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Storlie, C., & Helton, J. C. (2005). Multiple predictor smoothing methods for sensitivity analysis. In Proceedings of the 2005 Winter Simulation Conference (Vol. 2005, pp. 231-239). [1574256] https://doi.org/10.1109/WSC.2005.1574256

Multiple predictor smoothing methods for sensitivity analysis. / Storlie, Curtis; Helton, Jon C.

Proceedings of the 2005 Winter Simulation Conference. Vol. 2005 2005. p. 231-239 1574256.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Storlie, C & Helton, JC 2005, Multiple predictor smoothing methods for sensitivity analysis. in Proceedings of the 2005 Winter Simulation Conference. vol. 2005, 1574256, pp. 231-239, 2005 Winter Simulation Conference, Orlando, FL, United States, 12/4/05. https://doi.org/10.1109/WSC.2005.1574256
Storlie C, Helton JC. Multiple predictor smoothing methods for sensitivity analysis. In Proceedings of the 2005 Winter Simulation Conference. Vol. 2005. 2005. p. 231-239. 1574256 https://doi.org/10.1109/WSC.2005.1574256
Storlie, Curtis ; Helton, Jon C. / Multiple predictor smoothing methods for sensitivity analysis. Proceedings of the 2005 Winter Simulation Conference. Vol. 2005 2005. pp. 231-239
@inproceedings{1fcaee06d6cb4adab74139fd81fa2514,
title = "Multiple predictor smoothing methods for sensitivity analysis",
abstract = "The use of multiple predictor smoothing methods in sampling-based sensitivity analyses of complex models is investigated. Specifically, sensitivity analysis procedures based on smoothing methods employing the stepwise application of the following nonparametric regression techniques are described: (i) locally weighted regression (LOESS), (ii) additive models (GAMs), (iii) projection pursuit regression (PP_REG), and (iv) recursive partitioning regression (RP_REG). The indicated procedures are illustrated with both simple test problems and results from a performance assessment for a radioactive waste disposal facility (i.e., the Waste Isolation Pilot Plant). As shown by the example illustrations, the use of smoothing procedures based on nonparametric regression techniques can yield more informative sensitivity analysis results than can be obtained with more traditional sensitivity analysis procedures based on linear regression, rank regression or response surface regression when nonlinear relationships between model inputs and model predictions are present.",
author = "Curtis Storlie and Helton, {Jon C.}",
year = "2005",
doi = "10.1109/WSC.2005.1574256",
language = "English (US)",
isbn = "0780395204",
volume = "2005",
pages = "231--239",
booktitle = "Proceedings of the 2005 Winter Simulation Conference",

}

TY - GEN

T1 - Multiple predictor smoothing methods for sensitivity analysis

AU - Storlie, Curtis

AU - Helton, Jon C.

PY - 2005

Y1 - 2005

N2 - The use of multiple predictor smoothing methods in sampling-based sensitivity analyses of complex models is investigated. Specifically, sensitivity analysis procedures based on smoothing methods employing the stepwise application of the following nonparametric regression techniques are described: (i) locally weighted regression (LOESS), (ii) additive models (GAMs), (iii) projection pursuit regression (PP_REG), and (iv) recursive partitioning regression (RP_REG). The indicated procedures are illustrated with both simple test problems and results from a performance assessment for a radioactive waste disposal facility (i.e., the Waste Isolation Pilot Plant). As shown by the example illustrations, the use of smoothing procedures based on nonparametric regression techniques can yield more informative sensitivity analysis results than can be obtained with more traditional sensitivity analysis procedures based on linear regression, rank regression or response surface regression when nonlinear relationships between model inputs and model predictions are present.

AB - The use of multiple predictor smoothing methods in sampling-based sensitivity analyses of complex models is investigated. Specifically, sensitivity analysis procedures based on smoothing methods employing the stepwise application of the following nonparametric regression techniques are described: (i) locally weighted regression (LOESS), (ii) additive models (GAMs), (iii) projection pursuit regression (PP_REG), and (iv) recursive partitioning regression (RP_REG). The indicated procedures are illustrated with both simple test problems and results from a performance assessment for a radioactive waste disposal facility (i.e., the Waste Isolation Pilot Plant). As shown by the example illustrations, the use of smoothing procedures based on nonparametric regression techniques can yield more informative sensitivity analysis results than can be obtained with more traditional sensitivity analysis procedures based on linear regression, rank regression or response surface regression when nonlinear relationships between model inputs and model predictions are present.

UR - http://www.scopus.com/inward/record.url?scp=33846684615&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846684615&partnerID=8YFLogxK

U2 - 10.1109/WSC.2005.1574256

DO - 10.1109/WSC.2005.1574256

M3 - Conference contribution

AN - SCOPUS:33846684615

SN - 0780395204

SN - 9780780395206

VL - 2005

SP - 231

EP - 239

BT - Proceedings of the 2005 Winter Simulation Conference

ER -