Modeling and forecasting exchange rate volatility in Bangladesh using GARCH models: a comparison based on normal and Student’s t-error distribution

Modeling exchange rate volatility has remained crucially important because of its diverse implications. This study aimed to address the issue of error distribution assumption in modeling and forecasting exchange rate volatility between the Bangladeshi taka (BDT) and the US dollar ($). Using daily exchange rates for 7 years (January 1, 2008, to April 30, 2015), this study attempted to model dynamics following generalized autoregressive conditional heteroscedastic (GARCH), asymmetric power ARCH (APARCH), exponential generalized autoregressive conditional heteroscedstic (EGARCH), threshold generalized autoregressive conditional heteroscedstic (TGARCH), and integrated generalized autoregressive conditional heteroscedstic (IGARCH) processes under both normal and Student’s t -distribution assumptions for errors. It was found that, in contrast with the normal distribution, the application of Student’s t -distribution for errors helped the models satisfy the diagnostic tests and show improved forecasting accuracy. With such error distribution for out-of-sample volatility forecasting, AR(2)–GARCH(1, 1) is considered the best.