Modelling and Forecasting Volatility of Exchange Rates

This book investigates the volatility in the foreign exchange rate (forex) market using USD/MUR and EUR/MUR return series for the past ten years from October 2004 to October 2014. Both USD and EUR series exhibit volatility clustering and fat-tailed behaviours as shown by the results of the preliminary analysis. Furthermore, the presence of ARCH effects (conditional heteroscedasticity) in the residuals of both series validates the application of GARCH-type models for modelling the volatility of USD and EUR return series. Hence, the modelling proceeded with the use of the symmetric GARCH (1, 1) and asymmetric EGARCH (1, 1) models under two different innovations distribution (normal and Student-t). The factual results showed that the symmetric GARCH (1, 1) models are more appropriate to model the volatility of USD and EUR return series in the Mauritian Forex market. Moreover, GARCH (1, 1) with normal innovations and GARCH (1, 1) with Student-t innovations were deduced to be the best fitted model for USD and EUR return series respectively based on model selection criteria and model diagnostics results.