Using penalized contrasts for the change-point problem - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Reports Year : 2004

Using penalized contrasts for the change-point problem

Marc Lavielle

Abstract

A methodology for model selection based on a penalized contrast is developed. This methodology is applied to the change-point problem, for estimating the number of change points and their location. We aim to complete previous asymptotic results by constructing algorithms that can be used in diverse practical situations. First, we propose an adaptive choice of the penalty function for automatically estimating the dimension of the model, that is, the number of change points. In a Bayesian framework, we define the posterior distribution of the change-point sequence as a function of the penalized contrast. MCMC procedures are available for sampling this posterior distribution. The parameters of this distribution are estimated with a stochastic version of EM algorithm (SAEM). An application to EEG analysis and some Monte-Carlo experiments illustrate these algorithms.
Fichier principal
Vignette du fichier
RR-5339.pdf (282.93 Ko) Télécharger le fichier

Dates and versions

inria-00070662 , version 1 (19-05-2006)

Identifiers

  • HAL Id : inria-00070662 , version 1

Cite

Marc Lavielle. Using penalized contrasts for the change-point problem. RR-5339, INRIA. 2004, pp.18. ⟨inria-00070662⟩
151 View
1939 Download

Share

Gmail Facebook Twitter LinkedIn More