We present a modification of the well known transfer entropy (TE) which makes it able to detect, besides the direction and strength of the information transfer between coupled processes, its exact timing. The approach follows a decomposition strategy which identifies--according to a lag-specific formulation of the concept of Granger causality--the set of time delays carrying significant information, and then assigns to each of these delays an amount of information transfer such that the total contribution yields the overall TE. We propose also a procedure for the practical estimation from time series data of the relevant delays and lag-specific TE in both bivariate and multivariate settings. The proposed approach is tested in simulations and in real cardiovascular time series, showing the feasibility of lag-specific TE estimation, the ability to reflect expected mechanisms of cardiovascular regulation, and the necessity of using the multivariate TE to properly assess time-lagged information transfer in the presence of multiple interacting systems.

Faes, L., Nollo, G. (2013). Decomposing the transfer entropy to quantify lag-specific Granger causality in cardiovascular variability. In Poroceedings of the 35th Annual International Conference of the IEEE EMBS (pp.5049-5052) [10.1109/EMBC.2013.6610683].

Decomposing the transfer entropy to quantify lag-specific Granger causality in cardiovascular variability

Faes, Luca;
2013-01-01

Abstract

We present a modification of the well known transfer entropy (TE) which makes it able to detect, besides the direction and strength of the information transfer between coupled processes, its exact timing. The approach follows a decomposition strategy which identifies--according to a lag-specific formulation of the concept of Granger causality--the set of time delays carrying significant information, and then assigns to each of these delays an amount of information transfer such that the total contribution yields the overall TE. We propose also a procedure for the practical estimation from time series data of the relevant delays and lag-specific TE in both bivariate and multivariate settings. The proposed approach is tested in simulations and in real cardiovascular time series, showing the feasibility of lag-specific TE estimation, the ability to reflect expected mechanisms of cardiovascular regulation, and the necessity of using the multivariate TE to properly assess time-lagged information transfer in the presence of multiple interacting systems.
Settore ING-INF/06 - Bioingegneria Elettronica E Informatica
lug-2013
35th Annual International Conference of the IEEE EMBS
Osaka
2013
2013
4
Faes, L., Nollo, G. (2013). Decomposing the transfer entropy to quantify lag-specific Granger causality in cardiovascular variability. In Poroceedings of the 35th Annual International Conference of the IEEE EMBS (pp.5049-5052) [10.1109/EMBC.2013.6610683].
Proceedings (atti dei congressi)
Faes, Luca; Nollo, Giandomenico
File in questo prodotto:
File Dimensione Formato  
A26-IEEE-EMBS2013-Faes.pdf

Solo gestori archvio

Dimensione 364.47 kB
Formato Adobe PDF
364.47 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10447/276595
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 8
  • ???jsp.display-item.citation.isi??? 7
social impact