Article Dans Une Revue Journal of Applied Statistics Année : 2012

A new stochastic process to model Heart Rate series during exhaustive run and an estimator of its fractality parameter

Résumé

In order to interpret and explain the physiological signal behaviors, it can be interesting to find some constants among the fluctuations of these data during all the effort or during different stages of the race (which can be detected using a change points detection method). Several recent papers have proposed the long-range dependence (Hurst) parameter as such a constant. However, their results induce two main problems. Firstly, DFA method is usually applied for estimating this parameter. Clearly, such a method does not provide the most efficient estimator and moreover it is not at all robust even in the case of smooth trends. Secondly, this method often gives estimated Hurst parameters larger than $1$, which is the larger possible value for long memory stationary processes. In this article we propose solutions for both these problems and we define a new model allowing such estimated parameters.

Fichier principal
Vignette du fichier
JRSS.pdf (867.42 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence
Loading...

Dates et versions

hal-00176298 , version 1 (03-10-2007)
hal-00176298 , version 2 (26-03-2008)

Licence

Identifiants

Citer

Jean-Marc Bardet, Véronique Billat, Imen Kammoun. A new stochastic process to model Heart Rate series during exhaustive run and an estimator of its fractality parameter. Journal of Applied Statistics, 2012, 39 (6), pp.1331-1351. ⟨hal-00176298v2⟩
289 Consultations
634 Téléchargements

Altmetric

Partager

  • More