Arima-garch Modeling of Monthly Crude Oil Prices Volatility from Nigeria
Eke, Charles Ngome *
Department of Mathematics and Statistics, Federal Polytechnic Nekede, Owerri, Imo State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
This research work, studied the hybrid of autoregressive integrated moving average (ARIMA) and generalized autoregressive conditional heteroscedasticity models that best fit monthly crude oil price volatility of Nigeria between January, 2010 to March, 2021. The study collected secondary data from quarterly Central Bank of Nigeria (CBN) Statistical Bulletin, June, 2021 on monthly crude oil price of Nigeria to compute the monthly crude oil price returns. The ARIMA-GARCH modeling was adopted for this work. The series was tested for stationarity using Augmented Dickey Fuller test. Several ARIMA -GARCH models were applied to the monthly crude oil price returns to ascertain the best fit model for the series. The ARIMA (2, 0, 5)-GARCH(1,4) model was selected as the best fit for the data since it has minimum values of Akaike Information Criteria and Mean Squared Errors. The forecasted period showed a crude oil price with an unstable monthly crude oil price returns. Therefore, the government of Nigeria was advised to be conservative when planning with revenue from crude oil sales in future.
Keywords: Crude Oil, ARIMA, GARCH, model adequacy, forecast