Variance-based sensitivity analysis using harmonic analysis

Jean-Yves Tissot 1, 2 Clémentine Prieur 1, 2
2 MOISE - Modelling, Observations, Identification for Environmental Sciences
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : Fourier Amplitude Sensitivity Test (FAST) and Random Balance Design (RBD) are popular methods of estimating variance-based sensitivity indices. We revisit them in light of the discrete Fourier transform (DFT) on finite subgroups of the torus and randomized orthogonal array sampling. We then study the estimation error of both these methods. This allows to improve FAST and to derive explicit rates of convergence of its estimators by using the framework of lattice rules. We also give a natural generalization of the classic RBD by using randomized orthogonal arrays having any parameters, and we provide a bias correction method for its estimators.
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00680725
Contributor : Jean-Yves Tissot <>
Submitted on : Tuesday, March 20, 2012 - 9:27:53 AM
Last modification on : Wednesday, April 11, 2018 - 1:58:38 AM
Document(s) archivé(s) le : Thursday, June 21, 2012 - 2:23:08 AM

File

FAST_RBD_revisited.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00680725, version 1

Collections

Citation

Jean-Yves Tissot, Clémentine Prieur. Variance-based sensitivity analysis using harmonic analysis. 2012. ⟨hal-00680725⟩

Share

Metrics

Record views

865

Files downloads

198