Performance of symmetric and asymmetric GARCH models in Nigeria's crude oil markets

Abstract

The study investigated the performance of symmetric and asymmetric GARCH models in Nigerian crude oil markets between 1999 and 2023. The models adopted for the study were the symmetric and asymmetric GARCH Models. The data for the study was the crude oil price in Naira/Dollar from January 1999 to April 2023 sourced for and extracted from the Central Bank of Nigeria (CBN) Statistical Database. The results of the estimation of the models show that an expected non-stationary process exists in price from the time plot of the raw series and returns on Nigerian crude oil price. The series shows continuous trending across the time of the observations (data series) on the vertical axis while the time is on the horizontal axis. The results show that returns on Nigerian crude oil prices using the symmetric and asymmetric GARCH Models confirmed TGARCH as appropriate under the student-t with fixed parameter degree of freedom (df=10) as the model of best fit These were considered based on model performance evaluation using the Akaike information criteria. The summary of the diagnostic test confirmed TGARCH as adequate in modelling Nigeria's crude oil price. This indicates that the shocks associated with Nigerian crude oil prices persist over a long time. In addition, the long memory hypothesis model using the Nigerian crude oil price is stronger and so, this shows evolving situations that portray that those markets are not stable. The study recommended amongst others that the TGARCH model can capture a large impact on volatility, forcing investors and traders to incorporate risk into their market strategies.