Compensated transfer entropy as a tool for reliably estimating information transfer in physiological time series

Luca Faes, Giandomenico Nollo, Alberto Porta

Research output: Contribution to journalArticle

Abstract

We present a framework for the estimation of transfer entropy (TE) under the conditions typical of physiological system analysis, featuring short multivariate time series and the presence of instantaneous causality (IC). The framework is based on recognizing that TE can be interpreted as the difference between two conditional entropy (CE) terms, and builds on an efficient CE estimator that compensates for the bias occurring for high dimensional conditioning vectors and follows a sequential embedding procedure whereby the conditioning vectors are formed progressively according to a criterion for CE minimization. The issue of IC is faced accounting for zero-lag interactions according to two alternative empirical strategies: if IC is deemed as physiologically meaningful, zero-lag effects are assimilated to lagged effects to make them causally relevant; if not, zero-lag effects are incorporated in both CE terms to obtain a compensation. The resulting compensated TE (cTE) estimator is tested on simulated time series, showing that its utilization improves sensitivity (from 61% to 96%) and specificity (from 5/6 to 0/6 false positives) in the detection of information transfer respectively when instantaneous effect are causally meaningful and non-meaningful. Then, it is evaluated on examples of cardiovascular and neurological time series, supporting the feasibility of the proposed framework for the investigation of physiological mechanisms.

Original languageEnglish
Pages (from-to)198-219
Number of pages22
JournalEntropy
Volume15
Issue number1
DOIs
Publication statusPublished - Jan 2013

Fingerprint

information transfer
estimating
entropy
time lag
conditioning
estimators
systems analysis
embedding
optimization
sensitivity

Keywords

  • Cardiovascular variability
  • Conditional entropy
  • Instantaneous causality
  • Magnetoencephalography
  • Time delay embedding

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

Compensated transfer entropy as a tool for reliably estimating information transfer in physiological time series. / Faes, Luca; Nollo, Giandomenico; Porta, Alberto.

In: Entropy, Vol. 15, No. 1, 01.2013, p. 198-219.

Research output: Contribution to journalArticle

@article{64bfd5c73cd34a7fab8e3531d114a33a,
title = "Compensated transfer entropy as a tool for reliably estimating information transfer in physiological time series",
abstract = "We present a framework for the estimation of transfer entropy (TE) under the conditions typical of physiological system analysis, featuring short multivariate time series and the presence of instantaneous causality (IC). The framework is based on recognizing that TE can be interpreted as the difference between two conditional entropy (CE) terms, and builds on an efficient CE estimator that compensates for the bias occurring for high dimensional conditioning vectors and follows a sequential embedding procedure whereby the conditioning vectors are formed progressively according to a criterion for CE minimization. The issue of IC is faced accounting for zero-lag interactions according to two alternative empirical strategies: if IC is deemed as physiologically meaningful, zero-lag effects are assimilated to lagged effects to make them causally relevant; if not, zero-lag effects are incorporated in both CE terms to obtain a compensation. The resulting compensated TE (cTE) estimator is tested on simulated time series, showing that its utilization improves sensitivity (from 61{\%} to 96{\%}) and specificity (from 5/6 to 0/6 false positives) in the detection of information transfer respectively when instantaneous effect are causally meaningful and non-meaningful. Then, it is evaluated on examples of cardiovascular and neurological time series, supporting the feasibility of the proposed framework for the investigation of physiological mechanisms.",
keywords = "Cardiovascular variability, Conditional entropy, Instantaneous causality, Magnetoencephalography, Time delay embedding",
author = "Luca Faes and Giandomenico Nollo and Alberto Porta",
year = "2013",
month = "1",
doi = "10.3390/e15010198",
language = "English",
volume = "15",
pages = "198--219",
journal = "Entropy",
issn = "1099-4300",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "1",

}

TY - JOUR

T1 - Compensated transfer entropy as a tool for reliably estimating information transfer in physiological time series

AU - Faes, Luca

AU - Nollo, Giandomenico

AU - Porta, Alberto

PY - 2013/1

Y1 - 2013/1

N2 - We present a framework for the estimation of transfer entropy (TE) under the conditions typical of physiological system analysis, featuring short multivariate time series and the presence of instantaneous causality (IC). The framework is based on recognizing that TE can be interpreted as the difference between two conditional entropy (CE) terms, and builds on an efficient CE estimator that compensates for the bias occurring for high dimensional conditioning vectors and follows a sequential embedding procedure whereby the conditioning vectors are formed progressively according to a criterion for CE minimization. The issue of IC is faced accounting for zero-lag interactions according to two alternative empirical strategies: if IC is deemed as physiologically meaningful, zero-lag effects are assimilated to lagged effects to make them causally relevant; if not, zero-lag effects are incorporated in both CE terms to obtain a compensation. The resulting compensated TE (cTE) estimator is tested on simulated time series, showing that its utilization improves sensitivity (from 61% to 96%) and specificity (from 5/6 to 0/6 false positives) in the detection of information transfer respectively when instantaneous effect are causally meaningful and non-meaningful. Then, it is evaluated on examples of cardiovascular and neurological time series, supporting the feasibility of the proposed framework for the investigation of physiological mechanisms.

AB - We present a framework for the estimation of transfer entropy (TE) under the conditions typical of physiological system analysis, featuring short multivariate time series and the presence of instantaneous causality (IC). The framework is based on recognizing that TE can be interpreted as the difference between two conditional entropy (CE) terms, and builds on an efficient CE estimator that compensates for the bias occurring for high dimensional conditioning vectors and follows a sequential embedding procedure whereby the conditioning vectors are formed progressively according to a criterion for CE minimization. The issue of IC is faced accounting for zero-lag interactions according to two alternative empirical strategies: if IC is deemed as physiologically meaningful, zero-lag effects are assimilated to lagged effects to make them causally relevant; if not, zero-lag effects are incorporated in both CE terms to obtain a compensation. The resulting compensated TE (cTE) estimator is tested on simulated time series, showing that its utilization improves sensitivity (from 61% to 96%) and specificity (from 5/6 to 0/6 false positives) in the detection of information transfer respectively when instantaneous effect are causally meaningful and non-meaningful. Then, it is evaluated on examples of cardiovascular and neurological time series, supporting the feasibility of the proposed framework for the investigation of physiological mechanisms.

KW - Cardiovascular variability

KW - Conditional entropy

KW - Instantaneous causality

KW - Magnetoencephalography

KW - Time delay embedding

UR - http://www.scopus.com/inward/record.url?scp=84873185183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873185183&partnerID=8YFLogxK

U2 - 10.3390/e15010198

DO - 10.3390/e15010198

M3 - Article

AN - SCOPUS:84873185183

VL - 15

SP - 198

EP - 219

JO - Entropy

JF - Entropy

SN - 1099-4300

IS - 1

ER -