Abstract:
The analysis of financial crisis features is fundamental for the systemic risk assessment and prevention of events in the future. This study, proposes the use of an Evolutionary Game Theory, a novel approach that has no antecedents in the analysis of financial distress, in order to identify clusters (called dominant sets) of strongly interconnected firms in financial networks. The members of these dominant set are characterized by their high exposure to contagion risk, thus their identification will be indubitably useful to orientate financial policies. Moreover, the sequential analysis of the networks generates a set of financial market dislocation indicators that can be used as an alternative to the network density, a measure of systemic risk recently proposed in the literature. The methodology was applied on the analysis of a dynamic dataset of Pairwise Granger-causality networks constructed with some asset returns of the European Stock Market, where the indicators obtained based on dominant sets found revealed to be effective in predicting banking crises and describing the financial stability conditions in the Euro area.
