Uncertainty and specification challenges in exchange rates modeling under Bayesian model averaging

Traditional econometric models often struggle to capture the complexities and uncertainties inherent in exchange rate movements. This study investigates the dynamic relationship between the Nigerian Naira exchange rate and key economic variables using Bayesian Model Averaging (BMA), offering a robust framework to address model uncertainty and specification challenges. By integrating multiple predictors, BMA enhances forecasting accuracy, providing valuable insights for policymakers, investors and businesses navigating Nigeria's volatile economic landscape. Analysing exchange rate movements from 2007 to 2021, the study assesses BMA's effectiveness compared to traditional models, particularly in capturing non-linear relationships and time-varying volatility. Findings reveal that capital expenditure and the money supply are the most significant determinants of exchange rate fluctuations: capital expenditure negatively affects the exchange rate, while increased money supply leads to currency appreciation. These results highlight the potential of BMA to refine economic forecasts and improve decision-making in Nigeria's financial and policy sectors.