Effective network inference through multivariate information transfer estimation

Research output: Contribution to journalArticle

Abstract

Network representation has steadily gained in popularity over the past decades. In many disciplines such as finance, genetics, neuroscience or human travel to cite a few, the network may not directly be observable and needs to be inferred from time-series data, leading to the issue of separating direct interactions between two entities forming the network from indirect interactions coming through its remaining part. Drawing on recent contributions proposing strategies to deal with this problem such as the so-called “global silencing” approach of Barzel and Barabasi or “network deconvolution” of Feizi et al. (2013), we propose a novel methodology to infer an effective network structure from multivariate conditional information transfers. Its core principal is to test the information transfer between two nodes through a step-wise approach by conditioning the transfer for each pair on a specific set of relevant nodes as identified by our algorithm from the rest of the network. The methodology is model free and can be applied to high-dimensional networks with both inter-lag and intra-lag relationships. It outperforms state-of-the-art approaches for eliminating the redundancies and more generally retrieving simulated artificial networks in our Monte-Carlo experiments. We apply the method to stock market data at different frequencies (15 min, 1 h, 1 day) to retrieve the network of US largest financial institutions and then document how bank's centrality measurements relate to bank's systemic vulnerability.

Original languageEnglish
Pages (from-to)376-394
Number of pages19
JournalPhysica A: Statistical Mechanics and its Applications
Volume499
DOIs
Publication statusPublished - 1 Jun 2018

Fingerprint

information transfer
Information Transfer
inference
time lag
methodology
finance
neurology
Neuroscience
Methodology
Monte Carlo Experiment
Centrality
Deconvolution
vulnerability
Vertex of a graph
Time Series Data
Stock Market
Finance
Vulnerability
Interaction
Network Structure

Keywords

  • Bank network
  • Effective network
  • Indirect link
  • Systemic risk

Cite this

@article{4710d422c1ed4191bdeee81c14c9fe27,
title = "Effective network inference through multivariate information transfer estimation",
abstract = "Network representation has steadily gained in popularity over the past decades. In many disciplines such as finance, genetics, neuroscience or human travel to cite a few, the network may not directly be observable and needs to be inferred from time-series data, leading to the issue of separating direct interactions between two entities forming the network from indirect interactions coming through its remaining part. Drawing on recent contributions proposing strategies to deal with this problem such as the so-called “global silencing” approach of Barzel and Barabasi or “network deconvolution” of Feizi et al. (2013), we propose a novel methodology to infer an effective network structure from multivariate conditional information transfers. Its core principal is to test the information transfer between two nodes through a step-wise approach by conditioning the transfer for each pair on a specific set of relevant nodes as identified by our algorithm from the rest of the network. The methodology is model free and can be applied to high-dimensional networks with both inter-lag and intra-lag relationships. It outperforms state-of-the-art approaches for eliminating the redundancies and more generally retrieving simulated artificial networks in our Monte-Carlo experiments. We apply the method to stock market data at different frequencies (15 min, 1 h, 1 day) to retrieve the network of US largest financial institutions and then document how bank's centrality measurements relate to bank's systemic vulnerability.",
keywords = "Bank network, Effective network, Indirect link, Systemic risk",
author = "Dahlqvist, {Carl Henrik} and Jean-Yves Gnabo",
year = "2018",
month = "6",
day = "1",
doi = "10.1016/j.physa.2018.02.053",
language = "English",
volume = "499",
pages = "376--394",
journal = "Physica A: Statistical Mechanics and its Applications",
issn = "0378-4371",
publisher = "Elsevier",

}

TY - JOUR

T1 - Effective network inference through multivariate information transfer estimation

AU - Dahlqvist, Carl Henrik

AU - Gnabo, Jean-Yves

PY - 2018/6/1

Y1 - 2018/6/1

N2 - Network representation has steadily gained in popularity over the past decades. In many disciplines such as finance, genetics, neuroscience or human travel to cite a few, the network may not directly be observable and needs to be inferred from time-series data, leading to the issue of separating direct interactions between two entities forming the network from indirect interactions coming through its remaining part. Drawing on recent contributions proposing strategies to deal with this problem such as the so-called “global silencing” approach of Barzel and Barabasi or “network deconvolution” of Feizi et al. (2013), we propose a novel methodology to infer an effective network structure from multivariate conditional information transfers. Its core principal is to test the information transfer between two nodes through a step-wise approach by conditioning the transfer for each pair on a specific set of relevant nodes as identified by our algorithm from the rest of the network. The methodology is model free and can be applied to high-dimensional networks with both inter-lag and intra-lag relationships. It outperforms state-of-the-art approaches for eliminating the redundancies and more generally retrieving simulated artificial networks in our Monte-Carlo experiments. We apply the method to stock market data at different frequencies (15 min, 1 h, 1 day) to retrieve the network of US largest financial institutions and then document how bank's centrality measurements relate to bank's systemic vulnerability.

AB - Network representation has steadily gained in popularity over the past decades. In many disciplines such as finance, genetics, neuroscience or human travel to cite a few, the network may not directly be observable and needs to be inferred from time-series data, leading to the issue of separating direct interactions between two entities forming the network from indirect interactions coming through its remaining part. Drawing on recent contributions proposing strategies to deal with this problem such as the so-called “global silencing” approach of Barzel and Barabasi or “network deconvolution” of Feizi et al. (2013), we propose a novel methodology to infer an effective network structure from multivariate conditional information transfers. Its core principal is to test the information transfer between two nodes through a step-wise approach by conditioning the transfer for each pair on a specific set of relevant nodes as identified by our algorithm from the rest of the network. The methodology is model free and can be applied to high-dimensional networks with both inter-lag and intra-lag relationships. It outperforms state-of-the-art approaches for eliminating the redundancies and more generally retrieving simulated artificial networks in our Monte-Carlo experiments. We apply the method to stock market data at different frequencies (15 min, 1 h, 1 day) to retrieve the network of US largest financial institutions and then document how bank's centrality measurements relate to bank's systemic vulnerability.

KW - Bank network

KW - Effective network

KW - Indirect link

KW - Systemic risk

UR - http://www.scopus.com/inward/record.url?scp=85042367491&partnerID=8YFLogxK

U2 - 10.1016/j.physa.2018.02.053

DO - 10.1016/j.physa.2018.02.053

M3 - Article

VL - 499

SP - 376

EP - 394

JO - Physica A: Statistical Mechanics and its Applications

JF - Physica A: Statistical Mechanics and its Applications

SN - 0378-4371

ER -