Decomposing and backtesting a flexible specification for CoVaR

We introduce the Conditional Autoregressive Quantile-Located VaR (QL-CoCaViaR), that ex- tends the Conditional Value-at-Risk (Adrian and Brunnermeier, 2016) by using an estimation process capturing the state of joint distress of the financial system and of individual companies. Furthermore, we include autoregressive components of conditional quantiles to explicitly model volatility clustering and heteroskedasticity. We support our model with a large empirical analysis, in which we use both classical and novel backtesting methods. Our results show that the quantile- located relationships lead to relevant improvements in terms of predictive accuracy during stressed periods, providing a valuable tool for regulators to assess systemic events. © 2019 Elsevier B.V. All rights reserved.