Abstract: This paper investigates the dynamics of the Ghanaian Cedi (GHS) in relation to the US Dollar (USD) by applying a robust time series modeling technique. The study, which draws on monthly exchange rate data spanning from January 2015 to December 2024, applies the Autoregressive Integrated Moving Average (ARIMA) methodology to analyze and forecast currency movements. Stationarity was assessed using the Augmented Dickey-Fuller (ADF) test, followed by model identification through ACF and PACF plots. The best-fit model was selected using both the Akaike and Bayesian Information Criteria (AIC and BIC), with residual diagnostics ensuring model adequacy. The findings indicate that while the GHS/USD rate is non-stationary in its raw form, it becomes stationary after first differencing. The chosen ARIMA model shows strong predictive performance and captures the underlying trends and short-term fluctuations. The results lay emphasis on the influence of both internal and external macroeconomic variables on exchange rate volatility. The study offers implications for policymakers and investors by highlighting the value of time series modeling in traversing foreign exchange risks and informing economic planning.

Keywords: Exchange rate forecasting, ARIMA model, Ghanaian Cedi, time series analysis, monetary policy, exchange rate volatility

Pages: 165-172

WSEAS Transactions on