Bullion
Keywords
Inflation Rate, Regression with ARIMA error, ARIMA, Forecasting, Nigeria
Abstract
The paper aimed to determine the most appropriate method for forecasting inflation rate in Nigeria between Regression with Autoregressive Integrated Moving Average (ARIMA) errors and Autoregressive Integrated Moving Average (ARIMA) method. This study uses annual data for the period 1990-2021 from World Bank. One response variable and three predictor variables are adopted. The best ARIMA method was selected from the competing ARIMA methods using Akaike Information Criterion Corrected (AICC), while the root mean squared error (RMSE) and mean absolute percentage error (MAPE) are used to select the best forecast method between the two methods of interest. The result of this study showed that regression model with ARIMA (0,0,0) errors method is the best method for predicting inflation. The finding further revealed that the inflation rate in Nigeria will rise by 20.08 percent in 2030 and by 2040 it will reduce to 10.00 percent. This study suggests that the government should narrow down an Exchange policy that will help reduce fiscal gap, enhance government revenue, and bride the savings investment gap.
Issue
1
Volume
47
First Page
57
Last Page
64
Recommended Citation
Ugwuanyi, Barthelomew I.
(2023)
"Modeling and Forecasting Inflation Rate in Nigeria: A Comparison of Regression with ARIMA Errors and ARIMA Method,"
Bullion: Vol. 47:
No.
1, Article 5.
Available at:
https://dc.cbn.gov.ng/bullion/vol47/iss1/5