Modeling and Forecasting Volatility in USD/GBP Exchange Rate

Abstract

Rate changes can occur hourly, daily, or in large incremental shifts. These changes may impact firms by changing the cost of commodities imported from other countries and the demand for their goods among foreign consumers. Therefore, it is essential to forecast exchange rates to manage this business effect. This study aims to determine the best model for predicting volatility in the exchange rate between USD and GBP. In particular, we analyze exchange rates using the Autoregressive Integrated Moving Average (ARIMA) model and the volatility or variance model by Generalized Autoregressive Conditional Heteroscedasticity (GARCH). To determine the best model, the performance of each model is evaluated with several criteria, namely Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The results show that EGARCH(1,1) has the best forecasting performance in the out-sample section because it can better capture out-sample data patterns with minimum RMSE, MAE, and MAPE.