Modelling systemic risk using neural network quantile regression

Abstract We propose a novel approach to calibrate the conditional value-at-risk (CoVaR) of financial institutions based on neural network quantile regression. Building estimation results, we model systemic risk spillover effects in context across banks by considering marginal regression procedure. An out-of-sample analysis shows great performance compared linear baseline specification, signifying importance that nonlinearity plays for modelling risk. then three network-based measures from our fitted results. First, use Systemic Network Risk Index (SNRI) as measure total A comparison existing reveals offers new perspective due focus lower tail and allowance nonlinear effects. also introduce Fragility (SFI) Hazard (SHI) firm-specific measures, which allow us identify systemically relevant firms during crisis.