Volatility forecasting and risk management in some MENA stock markets: a nonlinear framework

In this paper, we estimate the value-at-risk (VaR) for some Middle East and North African emerging stock markets (Egypt, Israel, Turkey and Morocco) for the short and the long trading positions. We check whether considering for LM, asymmetries and fat-tails in the stock return's behaviour offer more accurate VaR forecasts. We compute the VaR for two ARCH/GARCH-type models including FIGARCH and FIAPARCH under two density functions: student and skewed student. The obtained results point out that that accounting for long dependence in return and volatility, fat-tails and asymmetry provides better one-day-ahead VaR forecasts. Furthermore, the FIAPARCH model out-performs the other models in the VaR forecasts. Finally, the FIAPARCH model provides for all the stock market indexes the lowest number of violations under the Basel II rules, given a risk exposure at the 99% confidence level. Our results offer potential implications for MENA stock markets risk quantifications, policy regulations and hedging strategies.