Value-at-Risk in turbulence time

Value-at-Risk (VaR) has been adopted as the cornerstone and common language of risk management by virtually all major financial institutions and regulators. However, this risk measure has failed to warn the market participants during the financial crisis. In this paper, we show this failure may come from the methodology that we use to calculate VaR and not necessarily for VaR measure itself. we compare two different methods for VaR calculation, 1. by assuming the normal distribution of portfolio return, 2. by using a bootstrap method in a nonparametric framework. The Empirical exercise is implemented on CAC40 index, and the results show us that the first method will underestimate the market risk - the failure of VaR measure occurs. Yet, the second method overcomes the shortcomings of the first method and provides results that pass the tests of VaR evaluation.