New AI clustering method maps financial network roles using ECB data
Egonet features reveal hidden roles like market intermediaries and cross-segment connectors.
Deep Dive
This paper proposes an interpretable role-based clustering approach for multi-layer financial networks. It constructs explainable node embeddings based on egonet features that capture direct and indirect trading relationships within and across market layers. Using transaction-level data from the ECB's Money Market Statistical Reporting (MMSR), the approach uncovers heterogeneous institutional roles such as market intermediaries, cross-segment connectors, and peripheral lenders or borrowers.
Key Points
- Method uses egonet features to build interpretable embeddings for financial institutions in multi-layer networks.
- Validated on ECB's MMSR transaction data, identifying roles like intermediaries, cross-segment connectors, and peripheral lenders.
- Clustering framework is modular—analysts can customize proximity measures, evaluation metrics, and clustering algorithms.
Why It Matters
Provides regulators with an interpretable method to detect systemic risk and functional dependencies in financial networks.