FORECASTING THE MALAYSIAN RINGGIT (MYR) EXCHANGE RATE: ARIMA VS GARCH

Abstract

The purpose of this study was to forecast the Malaysian Ringgit (MYR) exchange rate against the US Dollar (USD) and identify the optimal model for short-term forecasting purposes. Daily exchange rate data from December 1, 2003, to December 31, 2024, were analysed through the time series modeling technique known as Autoregressive Integrated Moving Average (ARIMA) and Generalized Autoregressive Conditional Heteroskedasticity (GARCH). The modelling process involved testing for stationarity, fitting the corresponding ARIMA and GARCH specifications, and evaluating model performance using Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE). Models were fitted with ARIMA (1,1,1), ARIMA (2,1,1), ARIMA (1,1,2), ARIMA (2,1,2), and GARCH (1,1). ARIMA (2,1,2) had the lowest error values and was accepted on MLR residual diagnostics. Using ARIMA (2,1,2), a five-day forecast and 95% confidence intervals were calculated, which included both point estimates and upper and lower bounds of future exchange rate values. The analysis reveals that the ARIMA (2,1,2) model outperformed the GARCH (1,1) model in shortterm predictions of the MYR/USD exchange rate. This highlights the model’s effectiveness for predicting future short-term currency movements. This study also emphasises the importance of appropriate data processing, thorough model testing, and careful model selection to forecast exchange rates. Future research could incorporate external economic data and explore more advanced or hybrid modelling methods to improve forecasting accuracy