REEXAMINING OF GARCH MODELS FOR EXCHANGE RATES VOLATILITY
This paper
compares the GARCH in Mean, the GARCH and the EGARCH
models in measuring exchange rate volatility to determine which model
is more efficient in terms of forecasting of volatility. Analysis of
forecasts of exchange rate volatility using Mean Squared Error (MSE)
shows that the EGARCH (l, l) model does better in describing the data for half
of the sample countries' exchange rates than the GARCH (1,1) and the GARCH-M
(1,1) models. When the Mean Absolute Percentage Error (MAPE) is used for
performance measure, the results are mixed. These results imply that the GARCH
(l, 1) model might not be an excellent model for measuring and forecasting
volatility when it varies over time.
___
- Baillie, R., T. Bollerslev (1989), "The Message in Daily Exchange Rates: Conditional Variance Tale," Journal ofBusiness & Economic Statistics, 7, 297-305.