TIME SERIES MODELING OF CRUDE OIL EXPORT IN NIGERIA: AN ITERATIVE APPROACH
Idongesit Michael Umoh & Ette Harrison Etuk
Department of Mathematics/Statistics, Ignatius Ajuru University of Education, Port Harcourt,
Department of Mathematics, Rivers State University, Port Harcourt,
idongesitumoh683@gmail.com ; ettetuk@yahoo.com, ettehetuk@gmail.com, etuk.ette@ust.edu.ng
ABSTRACT
It has been argued in recent times that the quantity of crude oil export increases as the year increases as well as production of crude oil, yet Nigerian economy decreases as the year increases. Hence, there is need to evaluate or build a model in order to help policy makers forecast the quantity of crude oil export in Nigeria. The aim of this paper is to fit an appropriate model to the crude oil export in Nigeria. Therefore, we estimated the autoregressive integrated moving averages: ARIMA(1,1,1), ARIMA(2,11), ARIMA(1,12), ARIMA(2,1,2) SARIMA (1,1,1) (1,1,1)12, SARIMA(1,1,1)(2,1,1)24 and their various autocorrelation function (ACF) and partial autocorrelation function (PACF). Based on Ljung-Box statistics, the ACF and PACF, we choose SARIMA(1,1,1)(2,1,1)24 as the appropriate model. This model has also, the least mean squared Error (MSE). Therefore we forecasted Nigeria’s crude oil export with SARIMA(1,1,1)(2,1,1)24; the results indicate that Nigeria crude oil export will decrease in the long run therefore there is need for diversification of Nigeria economy.
