MODELING AND FORCASTING OF CRUDE OIL PRICES UJSING THE BOX- JENKINS APPROACH:EVIDENCE FROM NIGERIA
Abstract
ABSTRACT
This study investigates the modeling and forecasting of daily crude oil prices using the Box-Jenkins approach, with data spanning from January 1,2023, to May 24,2024,sourced from the Central Bank of Nigeria (CBN). The objective was to develop an ARIMA model that accurately predicts future crude oil prices, providing insights into price dynamics crucial for stakeholders in the energy market. Descriptive statistics revealed a mean crude oil price of $84.87 per barrel and a standard deviation of
$5.62, with slight right skewness and moderate kurtosis. Initial stationarity testing using the Augmented Dickey-Fuller (ADF) test indicated non-stationarity,which was resolve through first differencing. The ACF and PACF plots of the differenced series suggested an ARIMA(5,1,2) model as the most appropriate, confirmed by the expert modeler in R software. The ARIMA(5,1,2) model parameters indicate significant short-term dependencies and effective error correction.Diagnostic checks showed that the residuals were approximately normally distributed and scattered around zero without any visible pattern. However,significant autocorrelation in the residuals, as indicated by the Ljung-Box test, suggested areas for model refinement. The 7-day ahead forecasts displayed initial fluctuations with a trend towards stabilization, indicating that while short-term volatility exists, pricesare expected to stabilize by the end of the forecast period. Based on these findings, the study recommends including relevant exogenous variables, and exploring alternative and hybrid models to enhance forecasting
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.