Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Multiple predictor smoothing methods for sensitivity analysis.

Technical Report ·
DOI:https://doi.org/10.2172/893126· OSTI ID:893126
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: (1) locally weighted regression (LOESS), (2) additive models, (3) projection pursuit regression, and (4) recursive partitioning regression. 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 quadratic regression when nonlinear relationships between model inputs and model predictions are present.
Research Organization:
Sandia National Laboratories
Sponsoring Organization:
USDOE
DOE Contract Number:
AC04-94AL85000
OSTI ID:
893126
Report Number(s):
SAND2006-4693
Country of Publication:
United States
Language:
English

Similar Records

Projection pursuit regression
Journal Article · Mon Nov 30 23:00:00 EST 1981 · J. Am. Stat. Assoc.; (United States) · OSTI ID:6232512

Projection pursuit regression
Conference · Thu Jan 31 23:00:00 EST 1980 · OSTI ID:5632461

Generalized smooth models
Conference · Thu Jul 23 00:00:00 EDT 1992 · OSTI ID:10157123