FORECASTING CONSUMER PRICE INDEX AND EXCHANGE RATE USING ARIMA MODELS: EMPIRICAL EVIDENCE FROM NIGERIA

Abstract

Considering the high level of uncertainty in the foreign exchange market and the adverse effects of inflation in Nigeria, the need to utilize current data to work on appropriate models capable of predicting the future values of exchange rates and CPI has become necessary to guide monetary policy makers. This paper applies the techniques of Autoregressive Integrated Moving Average models to forecast the CPI and exchange rates of Nigeria using the dataset of monthly CPI and Exchange rates of naira against US Dollar from January 2010 to August 2022 obtained from National Bureau of Statistics. After the original series were appropriately differenced to attain stationarity, autocorrelation function (ACF) and partial autocorrelation function (PACF) were used to select a number of tentative models for parameter estimation. Based on selection criteria such as AIC, SBIC, HQC, R² and Durbin Watson Statistics, ARIMA(1, 2, 0) was chosen as the best model for forecasting Nigeria’s monthly Inflation (CPI) and ARIMA(1, 1, 1) was selected as the most ideal model for forecasting monthly foreign exchange rates of Nigeria. The portmanteau tests carried out show that the residuals from both models are white noise which further confirms the adequacy of the fitted models. The results reveal that both inflation and Exchange rates of Naira against the Dollar will continue to rise. However, the rise in exchange rates for the short time is relatively steady. The findings from this study will furnish the monetary and policy makers with necessary information needed to reverse the expected trend