Scalable structural break detection

Tamas Elteto 1 Nikolaus Hansen 1 Cecile Germain-Renaud 1, 2 Pascal Bondon 3
1 TAO - Machine Learning and Optimisation
LRI - Laboratoire de Recherche en Informatique, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8623
3 Division Signaux
L2S - Laboratoire des signaux et systèmes
Abstract : This paper deals with a statistical model fitting procedure for non-stationary time series. This procedure selects the parameters of a piecewise autoregressive model using the Minimum Description Length principle. The existing chromosome representation of the piecewise autoregressive model and its corresponding optimisation algorithm are improved. First, we show that our proposed chromosome representation better captures the intrinsic properties of the piecewise autoregressive model. Second, we apply an optimisation algorithm, the Covariance Matrix Adaptation - Evolution Strategy, with which our setup converges faster to the optimal fit. Our proposed method achieves at least one order of magnitude performance improvement compared to the existing solution.
Document type :
Journal articles
Applied Soft Computing, Elsevier, 2012, <10.1016/j.asoc.2012.06.002>
Liste complète des métadonnées

https://hal.inria.fr/hal-00711843
Contributor : Cecile Germain-Renaud <>
Submitted on : Monday, June 25, 2012 - 8:38:19 PM
Last modification on : Thursday, February 9, 2017 - 4:01:33 PM
Document(s) archivé(s) le : Wednesday, September 26, 2012 - 3:05:45 AM

File

cmaes_for_segmentation_revisio...
Files produced by the author(s)

Identifiers

Citation

Tamas Elteto, Nikolaus Hansen, Cecile Germain-Renaud, Pascal Bondon. Scalable structural break detection. Applied Soft Computing, Elsevier, 2012, <10.1016/j.asoc.2012.06.002>. <hal-00711843>

Share

Metrics

Record views

817

Document downloads

312