Skip to Main content Skip to Navigation
Conference papers

Heuristique de pente pour les modèles de détection de ruptures multiples

Yann Guédon 1, 2, *
* Corresponding author
2 VIRTUAL PLANTS - Modeling plant morphogenesis at different scales, from genes to phenotype
CRISAM - Inria Sophia Antipolis - Méditerranée , INRA - Institut National de la Recherche Agronomique, UMR AGAP - Amélioration génétique et adaptation des plantes méditerranéennes et tropicales
Abstract : With regard to the retrospective multiple change-point detection problem, much effort has been devoted in recent years to the selection of the number of change points. But, the proposed approaches are either dedicated to specific models (e.g. Gaussian change in the mean model) or give unsatisfactory results for short or medium length sequences. We propose to apply the slope heuristic, a recently proposed non-asymptotic penalized likelihood criterion, for selecting the number of change points. We in particular apply the data-driven slope estimation method, the key point being to define a relevant penalty shape. The proposed approach is illustrated using two benchmark data sets.
Complete list of metadata

Cited literature [6 references]  Display  Hide  Download

https://hal.inria.fr/hal-01240298
Contributor : Christophe Godin <>
Submitted on : Wednesday, December 9, 2015 - 8:51:13 AM
Last modification on : Thursday, March 4, 2021 - 3:25:11 PM
Long-term archiving on: : Saturday, April 29, 2017 - 8:28:24 AM

File

submission_178.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01240298, version 1

Citation

Yann Guédon. Heuristique de pente pour les modèles de détection de ruptures multiples. 47èmes Journées de Statistique, Société Française de Statistique, Jun 2015, Lille, France. ⟨hal-01240298⟩

Share

Metrics

Record views

393

Files downloads

316