Robustness of the Sobol' indices to distributional uncertainty
Preprint
Abstract
Global sensitivity analysis (GSA) is used to quantify the influence of uncertain variables in a mathematical model. Prior to performing GSA, the user must specify (or implicitly assume), a probability distribution to model the uncertainty, and possibly statistical dependencies, of the variables. Determining this distribution is challenging in practice as the user has limited and imprecise knowledge of the uncertain variables. This article analyzes the robustness of the Sobol' indices, a commonly used tool in GSA, to changes in the distribution of the uncertain variables. A method for assessing such robustness is developed which requires minimal user specification and no additional evaluations of the model. Theoretical and computational aspects of the method are considered and illustrated through examples.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- On the consistency of Sobol indices with respect to stochastic ordering of model parametersESAIM: Probability and Statistics, 2019
- LCA of emerging technologies: addressing high uncertainty on inputs' variability when performing global sensitivity analysisScience of The Total Environment, 2017
- Sensitivity analysis of a sensitivity analysis: We are likely overlooking the impact of distributional assumptionsEcological Modelling, 2016
- Robust global sensitivity analysis under deep uncertainty via scenario analysisEnvironmental Modelling & Software, 2016
- Shapley Effects for Global Sensitivity Analysis: Theory and ComputationSIAM/ASA Journal on Uncertainty Quantification, 2016
- Scientific Ambiguity and Climate PolicyEnvironmental and Resource Economics, 2012
- Estimation of global sensitivity indices for models with dependent variablesComputer Physics Communications, 2012
- Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity indexComputer Physics Communications, 2010
- Theorems and examples on high dimensional model representationReliability Engineering & System Safety, 2003
- Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimatesMathematics and Computers in Simulation, 2001