Interaction uncertainty in financial networks

Paper: arXiv 2311.17875

Authors: Andrea Auconi

Abstract

A minimal stochastic dynamical model of the interbank network is introduced, with linear interactions mediated by an integral of recent variations. Defining stress as the variance over the banks’ states, the interaction correction to the stress expectation is derived and studied on the short-medium timescale in an expansion. It is shown that, while different interaction matrices can amplify or absorb fluctuations, on average interactions increase the stress expectation. More in general, this analytical framework enables to estimate the impact of uncertainty about financial exposures, and to draw conclusions about the importance of disclosure.

Complexity vs Empirical Score

• Math Complexity: 7.5/10
• Empirical Rigor: 3.0/10
• Quadrant: Lab Rats — theoretically deep, empirically untested

Why this score: The paper presents a sophisticated analytical framework using stochastic dynamical models and matrix expansions to study stress propagation in interbank networks. While highly novel and mathematically dense, it lacks real-world empirical validation or historical interbank data testing, relying primarily on theoretical derivations.

Research Flowchart

  flowchart TD
    A["Research Goal: Assess impact of interaction uncertainty & disclosures on interbank systemic risk"] --> B["Inputs: Interbank network matrix, bank states, & integral of recent variations"]
    B --> C["Model Construction: Build minimal stochastic dynamical model with linear interactions"]
    C --> D["Analytical Derivation: Perform stress variance expansion over short-medium timescales"]
    D --> E["Evaluation: Compare amplification vs. absorption across interaction matrices"]
    E --> F["Key Findings: Interactions increase average stress; disclosures reduce exposure uncertainty"]