Sensitivity analysis for multidimensional and functional outputs

Abstract : Let X:=(X1,…,Xp) be random objects (the inputs), defined on some probability space (Ω,F,ℙ) and valued in some measurable space E=E1×…×Ep. Further, let Y:=Y=f(X1,…,Xp) be the output. Here, f is a measurable function from E to some Hilbert space ℍ (ℍ could be either of finite or infinite dimension). In this work, we give a natural generalization of the Sobol indices (that are classically defined when Y∈ℝ), when the output belongs to ℍ. These indices have very nice properties. First, they are invariant under isometry and scaling. Further they can be, as in dimension 1, easily estimated by using the so-called Pick and Freeze method. We investigate the asymptotic behaviour of such an estimation scheme.
Type de document :
Article dans une revue
Electronic journal of statistics , Shaker Heights, OH : Institute of Mathematical Statistics, 2014, 8 (1), pp.575-603. 〈10.1214/14-EJS895〉
Domaine :

https://hal.archives-ouvertes.fr/hal-01980642
Contributeur : Fabrice Gamboa <>
Soumis le : lundi 14 janvier 2019 - 15:39:55
Dernière modification le : jeudi 7 février 2019 - 14:25:30

Citation

Fabrice Gamboa, Alexandre Janon, Thierry Klein, Agnès Lagnoux. Sensitivity analysis for multidimensional and functional outputs. Electronic journal of statistics , Shaker Heights, OH : Institute of Mathematical Statistics, 2014, 8 (1), pp.575-603. 〈10.1214/14-EJS895〉. 〈hal-01980642〉

Métriques

Consultations de la notice