De l’estimation à la prévision des crises financières : quelles implications pour l’allocation d’actifs ?

Dr Bokreta's project is to propose an index that evaluates the intensity of turbulences and then the potential future impacts on financial markets. This Index should capture the heterogeneity of investor horizons – from day traders to institutional investors and has the quality to be both simply computable and easy to understand.
It will be called the “New Index of Market Shocks (NIMS)” and is computed in three steps: volatility decomposition, volatility signal information synthetization and cumulative density function fit.
The objective is to identify regimes of financial crises that are defined when risk measure exceeds the arbitrary threshold corresponding to a 90% confidence level. The NIMS highest values during the period 2000-2003 mainly show that the French market is characterized by a strong instability over these years, whereas it is rather quite calm since 2003, with a rebirth of volatility in the recent months corresponding to the latest credit events. Moreover, the Index accounts for the multiscale features of market volatility and is proven to be robust to the distributional properties of data.