ANOVA kernels and RKHS of zero mean functions for model-based sensitivity analysis

Abstract : Given a reproducing kernel Hilbert space H of real-valued functions and a suitable measure mu over the source space D (subset of R), we decompose H as the sum of a subspace of centered functions for mu and its orthogonal in H. This decomposition leads to a special case of ANOVA kernels, for which the functional ANOVA representation of the best predictor can be elegantly derived, either in an interpolation or regularization framework. The proposed kernels appear to be particularly convenient for analyzing the e ffect of each (group of) variable(s) and computing sensitivity indices without recursivity.
Document type :
Journal articles
Liste complète des métadonnées

Cited literature [19 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00601472
Contributor : Nicolas Durrande <>
Submitted on : Friday, December 7, 2012 - 6:29:44 PM
Last modification on : Tuesday, October 23, 2018 - 2:36:08 PM
Document(s) archivé(s) le : Monday, March 11, 2013 - 11:56:07 AM

Files

ANOVA_kernels_for_SA.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00601472, version 2
  • ARXIV : 1106.3571

Citation

Nicolas Durrande, David Ginsbourger, Olivier Roustant, Laurent Carraro. ANOVA kernels and RKHS of zero mean functions for model-based sensitivity analysis. Journal of Multivariate Analysis, Elsevier, 2013, 115, pp.57-67. ⟨hal-00601472v2⟩

Share

Metrics

Record views

576

Files downloads

480