Forecasting Crude Oil Production Volumes in the Niger Delta Using ARIMA Time Series Models

Nigeria's crude oil production, predominantly sourced from the Niger Delta region in the South South geopolitical zone, is subject to significant volatility driven by pipeline vandalism, operational shutdowns, and global demand fluctuations. This study applies the Autoregressive Integrated Moving Average modelling framework to forecast monthly crude oil production volumes using data from the Nigerian Upstream Petroleum Regulatory Commission covering January 2015 to December 2023. The Box-Jenkins methodology is followed systematically: stationarity is tested using the Augmented Dickey-Fuller test, autocorrelation and partial autocorrelation functions are examined for order identification, and multiple candidate models are compared using the Akaike Information Criterion and Bayesian Information Criterion. The ARIMA (2,1,2) model is selected as the best-fitting specification, producing twelve-month ahead forecasts with a root mean square error of 48,000 barrels per day. Forecast intervals are computed at the 95 percent confidence level to quantify prediction uncertainty. The model identifies a downward structural trend in production volumes from 2020 onward, consistent with documented infrastructure deterioration in Bayelsa and Delta States. The study recommends integration of ARIMA forecasts into national budget planning frameworks to improve fiscal revenue projections. These results demonstrate the applicability of time series econometrics to energy sector analysis within the Nigerian context. Keywords: ARIMA, crude oil forecasting, Niger Delta, Box-Jenkins, time series analysis.