Guaranteed characterization of exact non-asymptotic confidence regions in nonlinear parameter estimation

Abstract : Recently, a new family of methods has been proposed for characterizing accuracy in nonlinear parameter estimation by Campi et al. These methods make it possible to obtain exact, non-asymptotic con dence regions for the parameter estimates under relatively mild assumptions on the noise distribution, namely that the noise samples are independently and symmetrically distributed. The numerical characterization of an exact con dence region with this new approach is far from being trivial, however. The aim of this paper is to show how interval analysis, which has been used for a guaranteed characterization of con dence regions for the parameter vector in other contexts, can contribute.
Complete list of metadatas

https://hal-supelec.archives-ouvertes.fr/hal-00819488
Contributor : Michel Kieffer <>
Submitted on : Wednesday, May 1, 2013 - 4:05:31 PM
Last modification on : Thursday, October 17, 2019 - 12:36:54 PM
Long-term archiving on : Friday, August 2, 2013 - 4:06:02 AM

File

Nolcos2013_KW.pdf
Files produced by the author(s)

Identifiers

Citation

Michel Kieffer, Eric Walter. Guaranteed characterization of exact non-asymptotic confidence regions in nonlinear parameter estimation. 9th IFAC Symposium on Nonlinear Control Systems (NOLCOS 2013), Sep 2013, Toulouse, France. pp.56-61, ⟨10.3182/20130904-3-FR-2041.00019⟩. ⟨hal-00819488⟩

Share

Metrics

Record views

575

Files downloads

686