Events and Meetings of Italian Statistical Society, Advances in Latent Variables - Methods, Models and Applications

Font Size: 
Bayesian quantile regression for tail risk interdependence
Mauro Bernardi, Ghislaine Gayraud, Lea Petrella

Last modified: 2013-06-14

Abstract


Recent financial disasters emphasized the need to investigate the consequence associated with the co-movement between the market and financial institutions’ assets; episodes of contagion among institutions are frequently observed and increase the probability of large capital losses. Commonly used risk management tools fail to account for potential spillover effects among institutions because they provide individual risk assessment. For this reason different systemic risk measures have been proposed in literature to analyse the interdependence effects of extremeevents. We contribute to this field by providing an estimation tool for evaluating the Conditional Value-at-Risk (CoVaR) defined as a risk measure of a financial institution conditioned on another financial institution being under distress. In particular our approach to estimate the CoVaR considers a quantile regression in a Bayesian inference framework.

Full Text: PDF