EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3979. Balance in financial signed networks: a new systemic risk measure

Invited abstract in session WD-63: Complexity and Financial Patterns, stream OR in Banking, Finance and Insurance: New Tools for Risk Management.

Wednesday, 14:30-16:00
Room: S14 (building: 101)

Authors (first author is the speaker)

1. Rosanna Grassi
Dep.t of Statistics and Quantitative Methods, Universita Milano-Bicocca
2. Paolo Bartesaghi
Dipartimento di Statistica e Metodi Quantitativi, University of Milano Bicocca
3. Fernando Diaz-Diaz
Institute of Cross Disciplinary Physics and Complex Systems
4. Pierpaolo Uberti
Università di Genova

Abstract

Changes in the correlation structures between asset returns strongly reflect the systemic events in the financial context. Commonly, correlation is used in absolute value. However, the question remains as to what information about the systemic risk is contained in the correlation signs, especially when interpreted as the weights of the edges in a signed network. From this perspective, balance theory can provide an interesting insight into catching signals of systemic crises and a meaningful tool to manage portfolio risk. The crucial point is that in balanced systems it is easier to predict the behaviour of a single entity. Hence, the balance contributes to increasing the level of "predictability" of the network behaviour. In a financial context, this is essential in managing risk to face systemic crises. We show how the global balance index can be interpreted as a possible indicator of systemic risk. Moreover, we compare the global balance with some systemic risk indicators from the literature. We focus on the indicators that well-capture the loss of linear independence in the matrix of observations through the analysis of the minimum eigenvalues of the correlation matrix, interpreted as the distance from an appropriate space of rank-defective matrices.
Preliminary results are promising and they create a new bridge between systemic risk measures and signed networks.

Keywords

Status: accepted

Back to the list of papers