An Entropy-Based Early Warning Indicator for Systemic Risk

Abstract

The purpose of this paper is the construction of an early warning indicator for systemic risk using entropy measures. The analysis is based on the cross-sectional distribution of marginal systemic risk measures such as Marginal Expected Shortfall, Delta CoVaR and network connectedness. These measures are conceived at a single institution for the financial industry in the Euro area. We estimate entropy on these measures by considering different definitions (Shannon, Tsallis and Renyi). Finally, we test if these entropy indicators show forecasting abilities in predicting banking crises. In this regard, we use the variable presented in Babeckỳ et al. (2012) and Alessi and Detken (2011) from European Central Bank.

Entropy indicators show promising forecast abilities to predict financial and banking crisis. The proposed early warning signals reveal to be effective in forecasting financial distress conditions.

Keywords: Entropy, systemic risk measures, early warning indicators, aggregation

JEL Classification: C10, C11, G12, G29