Online Monitoring of Financial Market Information-Flow Networks Under External Shocks: A Rolling Directed-ERGM and Control-Chart Framework
Open Access
- 12 March 2026
- journal article
- research article
- Published by MDPI AG in Mathematics
- Vol. 14 (6) , 961
Abstract
Amid frequent external shocks and deepening market linkages, the information-transmission structure of financial markets is more prone to phase-specific abrupt changes, creating a need for real-time monitoring methods. This study develops an online framework to track financial information-flow networks and to provide early warnings of structural changes under exogenous shocks. Methodologically, information-flow networks are constructed from return spillovers using the Diebold–Yilmaz framework. An Exponential Random Graph Model is then employed to quantify how exogenous variables affect edge formation. Statistical process control methods, namely the Multivariate Cumulative Sum (MCUSUM) and Multivariate Exponentially Weighted Moving Average (MEWMA), are introduced to online monitoring of exogenous-effect coefficients. The simulation study uses simulated data to assess whether the two charts are properly calibrated and sensitive to alarms. The empirical study uses Shanghai Stock Exchange (SSE) 180 constituent stocks and exogenous variables—7-day Fixing Repo Rate (FR007), M2 growth rate (M2), the China Economic Policy Uncertainty Index (CEPU), and the Global Economic Policy Uncertainty Index (GEPU) over 2011–2025. The results indicate that both charts achieve the target in-control average run length, and detection accelerates with shock magnitude; FR007 is generally negative, M2 is positive, and uncertainty measures vary strongly over time; monitoring reveals shock clustering and long-term drift, implying both shock amplification and structural drift in the information-flow network. Practically, the framework provides an implementable warning tool for tracking shock amplification, supporting timely risk management.Keywords
Funding Information
- Natural Science Foundation of Guangdong Province (2024A1515012587)
This publication has 87 references indexed in Scilit:
- Research on the diffusion effects of data value in supply chainsJournal of Asian Economics, 2025
- Concentration in cross‐border research collaborations and MNCs ' knowledge creation in a host countryStrategic Management Journal, 2025
- Bayesian autoregressive online change-point detection with time-varying parametersCommunications in Nonlinear Science and Numerical Simulation, 2024
- Monitoring the Dynamic Networks of Stock Returns with an Application to the Swedish Stock MarketComputational Economics, 2024
- Monetary policy shocks and the signaling channel of monetary policy in ChinaChina Economic Quarterly International, 2023
- Dynamic nonlinear connectedness between the green bonds, clean energy, and stock price: the impact of the COVID-19 pandemicAnnals of Operations Research, 2022
- Applications of Artificial Intelligence and Machine learning in smart citiesComputer Communications, 2020
- On Assessing the Performance of Sequential Procedures for Detecting a ChangeQuality and Reliability Engineering International, 2012
- Bid, ask and transaction prices in a specialist market with heterogeneously informed tradersJournal of Financial Economics, 1985
- Option pricing when underlying stock returns are discontinuousJournal of Financial Economics, 1976