hal-00665048, version 1
Asymptotic normality and efficiency of two Sobol index estimators
Alexandre Janon
a, 1, 2Thierry Klein
3Agnes Lagnoux-Renaudie
3Maëlle Nodet
a, 1Clémentine Prieur
a, 1, 2
Abstract: Many mathematical models involve input parameters, which are not precisely known. Global sensitivity analysis aims to identify the parameters whose uncertainty has the largest impact on the variability of a quantity of interest (output of the model). One of the statistical tools used to quantify the influence of each input variable on the output is the Sobol sensitivity index. We consider the statistical estimation of this index from a finite sample of model outputs: we present two estimators and state a central limit theorem for each. We show that one of these estimators has an optimal asymptotic variance. We also generalize our results to the case where the true output is not observable, and is replaced by a noisy version.
- a – Université Joseph Fourier - Grenoble I
- 1: MOISE (INRIA Grenoble Rhône-Alpes / LJK Laboratoire Jean Kuntzmann)
- CNRS : UMR5224 – INRIA – Laboratoire Jean Kuntzmann – Université Joseph Fourier - Grenoble I – Institut Polytechnique de Grenoble - Grenoble Institute of Technology
- 2: GdR MASCOT-NUM ((Méthodes d'Analyse Stochastique des Codes et Traitements Numériques))
- CNRS : GDR3179
- 3: Institut de Mathématiques de Toulouse (IMT)
- Université Paul Sabatier - Toulouse III – Université Toulouse le Mirail - Toulouse II – Université des Sciences Sociales - Toulouse I – Institut National des Sciences Appliquées de Toulouse – CNRS : UMR5219
- Domain : Mathematics/Statistics
Statistics/Applications
Statistics/Statistics Theory
Statistics/Computation - Keywords : sensitivity analysis – Sobol index – asymptotic normality – asymptotic efficiency – nonparametric Cramér-Rao bound – metamodel – response surface method – Kriging
- hal-00665048, version 1
- http://hal.inria.fr/hal-00665048
- oai:hal.inria.fr:hal-00665048
- From: Alexandre Janon
- Submitted on: Wednesday, 1 February 2012 10:32:48
- Updated on: Wednesday, 15 February 2012 15:16:05






Associated documents

Export