A Study on Sector-Wise Carbon Dioxide Emission Prediction Using Arima Model

Abstract

Timely prediction of Carbon Dioxide (CO₂) emissions is significant for climate change mitigation and sustainable environmental planning.  The study uses ARIMA model for forecasting the sectorial CO₂ emissions. Long-term CO₂ emissions were examined for the six different sectors: power, industry, transport, residential, domestic and international aviation industry based on historical CO₂ emissions dataset. The Augmented Dickey-Fuller test is performed to assess the stationarity of the time series dataset, and Autoregressive Integrated Moving Average (ARIMA) model is selected for CO₂ emission prediction. Several ARIMA (p, d, q) formations were examined for each sector, and the best suitable parameter are recognised and the model is evaluated based on various performance metrics including Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC).  The results shows the emission trends across various sectors and offers reliable statistical substance for forecasting future emissions. This study supports data-driven decision-making for environmental policy makers and contributes to the broader discourse on climate resilience and sustainable development.