Friedman Test Technique for Optimizing a Seasonal Box-Jenkins ARIMA Model Building

Abstract

This study employs the Box-Jenkins methodology for time series modelling to analyze Nigerian crude oil production data. To enhance the model-building process, the Friedman rank test for seasonality was utilized. Data were sourced from the Nigerian Petroleum Corporation’s Annual Statistical Bulletin, covering the period from January 1997 to September 2014. The findings indicate that the yearly and monthly mean distributions exhibit non-constant behavior over time, implying that both natural and governmental factors significantly impact crude oil production (COP) dynamics. Consequently, a seasonal ARIMA(0,1,1) (0,1,1)₁₂ model was fitted and deemed appropriate for the data.