PREDICTING PETROLEUM CONSUMPTION USING TRIGONOMETRIC REGRESSION MODEL

A. I. Taiwo; A. F. Adedotun; K. K. Adesanya; S. A. Agboluaje

A. I. Taiwo Department of Mathematical Sciences, Olabisi Onabanjo University, Ago Iwoye
A. F. Adedotun Covenant University, Ota
K. K. Adesanya Department of Health Records and Biostatistics, Ogun State College of Health Technology, Ilese-Ijebu
S. A. Agboluaje The Polytechnic Ibadan, Ibadan

Keywords: Energy and Oil Consumption, Prediction Accuracy, Regression, Forecasting, Time Series

Abstract

This study is used to model and forecast Nigerian motor gasoline consumption using the Trigonometric Regression model which has the capabilities of handling nonlinear time series. The time plot of Nigerian motor gasoline consumption showed the series is nonlinear. The Trigonometric regression model was estimated using the Ordinary Least Square method. From the result, the coefficients of the model influenced Nigerian motor gasoline consumption and a unit increase may lead to an increase or decrease. The values of coefficient of determination (R^2) revealed that the coefficients of the model explained the variations in Nigerian motor gasoline consumption up to 83%. The value of the adjusted coefficient of determination (R ̅^2 ) also revealed that the model is a good fit and has high predictive power. Therefore, the Nigerian motor gasoline consumption forecast from 1980 to 2038 indicated continuous fluctuations from year to year. The shape of the out-sample forecast from 2019 to 2038 exhibited a bell shape. Conclusively, based on the results obtained, the proposed model can be used to obtain future values for Nigerian motor gasoline consumption. This will enhance the Government and shareholders to put in place proper plans and logistics to curtail the challenges that may arise from Nigerian motor gasoline consumption and distribution presently and in the future.