Procyclicality of Empirical Measurements of Risk in Financial Markets

In our study we quantify the procyclicality of the historical Value at Risk (VaR). To do so we introduce a dynamic risk measure called Sample Quantile Process (SQP) – a rolling-window generalization of the VaR. By... [ view full abstract ]

In our study we quantify the procyclicality of the historical Value at Risk (VaR). To do so we introduce a dynamic risk measure called Sample Quantile Process (SQP) – a rolling-window generalization of the VaR. By conditioning the SQP on the current state of realized volatility, we are able to empirically quantify the procyclicality.
Moreover, using a simple GARCH(1,1) model, we conclude that this pro-cyclical effect is related to the clustering of volatility. At the same time, we prove that part of this procyclicality is intrinsically caused by the way the risk is estimated, using a historical VaR.