Time series modeling and forecasting of the demand of some petroleum products in Ghana

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
The study is an application of Box-Jenkins ARIMA modelling and forecasting to petroleum products demand in Ghana. Monthly data on demand levels from January, 1999 to December, 2010 of petroleum products namely; Gas Oil, Liquefied Petroleum Gas (LPG) and Premix Fuel were analysed and forecasts made 12 months ahead. Thus, after stationarity was established through differencing of the data, the sample ACF and the sample Partial Autocorrelation Function (PACF) of the differenced data were considered to generate possible models after which the AIC, AICc and BIC of the candidate models under the various data were examined and those candidate models with the smallest AIC, AICc and BIC were chosen as the best-fit models among the candidate models and used for forecasting. The best fit model for the National Gas Oil demand, National LPG demand and National Premix demand levels were found to be ARIMA(1,1,3), ARIMA(2,1,3) and SARIMA(3,1,0)(2,0,0)[12] respectively.
A Thesis submitted to the School of Graduate Studies, Kwame Nkrumah University of Science and Technology, Kumasi, in partial fulfilment of the requirements for the Degree of Master of Philosophy in Applied Mathematics