Transfer entropy between multivariate time series

Mao, Xuegeng; Shang, Pengjian

doi:10.1016/j.cnsns.2016.12.008

Cited by 72 publications

(26 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…. Q X X X X Γ represents the variance of the prediction error of 1 X by using the past of all signals except 2 X .…”

Section: B Granger Causalitymentioning

confidence: 99%

“…For each value of p, the box plots are displayed for the information flow from 1 to 2 (left-hand side) and from 2 to 1 (right-hand side). The null interval between the first and last quartiles obtained for the latter direction can be explained by the thresholding rule introduced to eliminate redundant predictors in kernel Granger method (this leads to a Granger index value equal to zero when none of the linear or nonlinear functions of 2 X implicitly introduced as features in the kernel approach is significantly correlated with the predicted variable).…”

Section: Physiology-based Modelmentioning

confidence: 99%

“…Specifically, if the variance of the linear AR prediction error of a first time series, called 1 X , at the present time is decreased by incorporating past measurements from a second time series, called 2 X , then 2 X can be said to Granger cause | . .…”

Section: B Granger Causalitymentioning

confidence: 99%

“…In that context, a variety of approaches have been developed in order to estimate effective connectivity and consequently identify the direction of causal interdependence. Those approaches include measures inspired from information theory such as transfer entropy (TE) [2], phase-related measures such as phase slope index (PSI) [3], and parametric-modeling-based measures such as Granger causality (GC) [4], [5]. TE is a nonlinear method characterized by model-free design (except for observation vectors dimensions) but needs computational capacity [6] whereas PSI is less complex but requires long data segments for available estimation [6].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Investigation of nonlinear granger causality in the context of epilepsy

Mahjoub¹,

Bellanger²,

Chaibi³

et al. 2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

show abstract

“…. Q X X X X Γ represents the variance of the prediction error of 1 X by using the past of all signals except 2 X .…”

Section: B Granger Causalitymentioning

confidence: 99%

Section: Physiology-based Modelmentioning

confidence: 99%

Section: B Granger Causalitymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Investigation of nonlinear granger causality in the context of epilepsy

Mahjoub¹,

Bellanger²,

Chaibi³

et al. 2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

show abstract

“…This allows the identification of the coupling between different parameters to derive and train descriptive and predictive models. Previous investigations focused on the definition of causality tests using time series (Paluš and Vejmelka, 2007), for instance using transfer entropy (Mao and Shang, 2017). But the understanding of causalities and the successful implementation of models requires a well-founded analysis of the influence of the sampled data 1 , and in some cases the inference of causality can be complicated by a bias in estimation from a limited amount of possibly noisy data (Paluš and Vejmelka, 2007).…”

Section: Introductionmentioning

confidence: 99%

Relative distances between homology groups to assess persistent defects in time series

Ochoa¹

2019

Preprint

View full text Add to dashboard Cite

It is common to consider a data-intensive strategy to be an appropriate way to develop systemic analyses in biology and physiology. Therefore, options for data collection, sampling, standardization, visualization, and interpretation determine how causes are identified in time series to build mathematical models. However, there are often biases in the collected data that can affect the validity of the model: while collecting enough large datasets seems to be a good strategy for reducing the bias of the collected data, persistent and dynamical anomalies in the data structure can affect the overall validity of the model. In this work we present a methodology based on the definition of homological groups to evaluate persistent anomalies in the structure of the sampled time series. In this evaluation relevant patterns in the combination of different time series are clustered and grouped to customize the identification of causal relationships between parameters. We test this methodology on data collected from patients using mobile sensors to test the response to physical exercise in real-world conditions and outside the lab. With this methodology we plan to obtain a patient stratification of the time series to customize models in medicine.

show abstract

On the Origins of the Intercorrelations Between Solar Wind Variables

Borovsky

2018

JGR Space Physics

View full text Add to dashboard Cite

It is well known that the time variations of the diverse solar wind variables at 1 AU (e.g., solar wind speed, density, proton temperature, electron temperature, magnetic field strength, specific entropy, heavy‐ion charge‐state densities, and electron strahl intensity) are highly intercorrelated with each other. In correlation studies of the driving of the Earth's magnetosphere‐ionosphere‐thermosphere system by the solar wind, these solar wind intercorrelations make determining cause and effect very difficult. In this report analyses of solar wind spacecraft measurements and compressible‐fluid computer simulations are used to study the origins of the solar wind intercorrelations. Two causes are found: (1) synchronized changes in the values of the solar wind variables as the plasma types of the solar wind are switched by solar rotation and (2) dynamic interactions (compressions and rarefactions) in the solar wind between the Sun and the Earth. These findings provide an incremental increase in the understanding of how the Sun‐Earth system operates.

show abstract

Transfer entropy between multivariate time series

Cited by 72 publications

References 33 publications

Investigation of nonlinear granger causality in the context of epilepsy

Investigation of nonlinear granger causality in the context of epilepsy

Relative distances between homology groups to assess persistent defects in time series

On the Origins of the Intercorrelations Between Solar Wind Variables

Contact Info

Product

Resources

About