Modeling and Forecasting Exchange Rate Values between Naira and US Dollar to Assess the Effect of COVID-19 Pandemic Period on the Rate

Abstract

Aims: To model exchange rate values between Naira and US Dollar in order to assess the effect of the COVID-19 pandemic period on the rate by examining the forecasts. Study Design: The study design is the longitudinal research design. Place and Duration of Study: Real life data of monthly average Nigerian Naira-US Dollar exchange rates from Jan.1991 to April 2020 obtained from the Data and Statistics publication of Central Bank of Nigeria. Methodology: The traditional time series ARIMA model is employed for forecasting Naira/Dollar exchange rate using Monthly data covering the period January, 1991 to April, 2020. The ACF and PACF plots showed tendency that the differenced exchange rate data behave as both having autoregressive and moving average processes as the ACF has significant peaks at different lags and a gradual decay to zero is indicated by the PACF. Results: Using the AIC, BIC, and HQC as model selection criterion the ARIMA (2, 1, 3) were selected as the model with the best fit for the exchange rate data as compared to the other selected models. The ACF of the residuals is plotted with lags up to 20 and the residuals coefficient did not exceed the 95% confidence limit which indicates that the model is a good fit and is appropriate for the data. Plots of in-sample and out-of-sample forecast were also made. Conclusion: The out-of-sample forecast plot for period of 12 months revealed that naira will continue to depreciate on US dollar for the period forecasted with very high tendencies during the COVID-19 pandemic period.