Systemic Risk Identification, Modelling, Analysis, and Monitoring: An Integrated Approach

TL;DR

This paper proposes an integrated, multi-layer network approach using tensor models to evaluate and monitor systemic risk in financial systems, aiming to support regulation and policy decisions.

Contribution

It introduces an innovative tensorial multilayer network model for systemic risk analysis, incorporating multiple risk sources and financial data for comprehensive evaluation.

Findings

Developed a tensor-based multilayer network model for systemic risk

Linked economic data with interbank exposures in the model

Outlined a multi-model hypernetwork for scenario analysis

Abstract

Research capacity is critical in understanding systemic risk and informing new regulation. Banking regulation has not kept pace with all the complexities of financial innovation. The academic literature on systemic risk is rapidly expanding. The majority of papers analyse a single source or a consolidated source of risk and its effect. A fraction of publications quantify systemic risk measures or formulate penalties for systemically important financial institutions that are of practical regulatory relevance. The challenges facing systemic risk evaluation and regulation still persist, as the definition of systemic risk is somewhat unsettled and that affects attempts to provide solutions. Our understanding of systemic risk is evolving and the awareness of data relevance is rising gradually; this challenge is reflected in the focus of major international research initiatives. There is a consensus that the direct and indirect costs of a systemic crisis are enormous as opposed to preventing it, and that without regulation the externalities will not be prevented; but there is no consensus yet on the extent and detail of regulation, and research expectations are to facilitate the regulatory process. This report outlines an integrated approach for systemic risk evaluation based on multiple types of interbank exposures through innovative modelling approaches as tensorial multilayer networks, suggests how to relate underlying economic data and how to extend the network to cover financial market information. We reason about data requirements and time scale effects, and outline a multi-model hypernetwork of systemic risk knowledge as a scenario analysis and policy support tool. The argument is that logical steps forward would incorporate the range of risk sources and their interrelated effects as contributions towards an overall systemic risk indicator, would perform an integral analysis of ...