Developing an Early Warning System for Financial Networks: An Explainable Machine Learning Approach DOI Creative Commons
Daren Purnell, Amir H. Etemadi, John Kamp

et al.

Entropy, Journal Year: 2024, Volume and Issue: 26(9), P. 796 - 796

Published: Sept. 17, 2024

Identifying the influential variables that provide early warning of financial network instability is challenging, in part due to complexity system, uncertainty a failure, and nonlinear, time-varying relationships between participants. In this study, we introduce novel methodology select that, from data-driven statistical modeling perspective, represent these may indicate trending toward instability. We variable selection leverages Shapley values modified Borda counts, combination with machine learning methods, create an explainable linear model predict relationship value weights validate new approach data collected March 2023 Silicon Valley Bank Failure. The models produced using method successfully identified trend only 14 input out possible 3160. use parsimonious developed by has potential identify key stability indicators while also increasing transparency complex system.

Language: Английский

Self-organization of the stock exchange to the edge of a phase transition: empirical and theoretical studies DOI Creative Commons
Dmitriev Av, A.N. Lebedev, Vasily Kornilov

et al.

Frontiers in Physics, Journal Year: 2025, Volume and Issue: 12

Published: Jan. 22, 2025

Our study is based on the hypothesis that stock exchanges, being nonlinear, open and dissipative systems, are capable of self-organization to edge a phase transition. To empirically support hypothesis, we find segments in hourly volume series for 3,000 stocks publicly traded companies, corresponding time exchange’s stay We provide theoretical justification present phenomenological model exchange first-order transition second-order In model, controlling parameter entropy as measure uncertainty information about share public company, guided by which players make decision buy/sell it. The order determined number transactions company’s shares, i.e., stock’s volume. By applying statistical tests AUC metric, found most effective early warning measures from set investigated critical deceleration measures, multifractal reconstructed space measures. practical significance our possibility exchanges can be extended with high frequency data future research.

Language: Английский

Citations

0
Developing an Early Warning System for Financial Networks: An Explainable Machine Learning Approach DOI Creative Commons
Daren Purnell, Amir H. Etemadi, John Kamp

et al.

Entropy, Journal Year: 2024, Volume and Issue: 26(9), P. 796 - 796

Published: Sept. 17, 2024

Identifying the influential variables that provide early warning of financial network instability is challenging, in part due to complexity system, uncertainty a failure, and nonlinear, time-varying relationships between participants. In this study, we introduce novel methodology select that, from data-driven statistical modeling perspective, represent these may indicate trending toward instability. We variable selection leverages Shapley values modified Borda counts, combination with machine learning methods, create an explainable linear model predict relationship value weights validate new approach data collected March 2023 Silicon Valley Bank Failure. The models produced using method successfully identified trend only 14 input out possible 3160. use parsimonious developed by has potential identify key stability indicators while also increasing transparency complex system.

Language: Английский

Citations

0