Publication

Abstract :

Greenhouse gas emission is a significant aggravator of global warming. Predicting CO2 emissions holds several advantages crucial for informed decision-making and sustainable resource management. Prediction and control of emissions are thus becoming increasingly important. Carbon emission prediction has been attempted using multiple paradigms. This work aims to employ a data-driven approach to produce highly accurate predictions of carbon emissions. Emission data from over 200 countries is treated as a time series and the Dynamic Mode Decomposition (DMD) algorithm is used to make predictions. The data-driven nature of the DMD algorithm can capture the underlying dynamics of the time series data and an accurate prediction can be done for the future period. This approach is analyzed and compared with current techniques to identify its benefits.