Development of regression models to forecast the CO2 emissions from fossil fuels in the BRICS and MINT countries

Abstract

This study focuses to propose regression models to forecast the fossil fuel-related carbon dioxide (FFCO2) emissions in the BRICS (Brazil, the Russian Federation, India, China, South Africa) and MINT (Mexico, Indonesia, Nigeria, Turkey) countries. In order to get an overview, a short review on the current status of the fossil fuel use and FFCO2 emissions is firstly presented. Then, statistical regression models are developed and verified by various statistical tests including coefficient of determination, F-and t-tests, and the actual versus predicted data plots. Additionally, the most significant variables affecting the FFCO2 emissions of the BRICS and MINT countries are statistically determined. Moreover, the forecasting accuracies and effectiveness of the proposed models are measured using the mean absolute percentage error. Using the proposed models, furthermore, the future FFCO2 emissions of the BRICS and MINT countries are estimated and discussed. The results show that BRICS and MINT countries will come to a position that will guide the world economy in the near future due to their significant superiorities. The modeling results reveal that the proposed models have a strong practical applicability potential. Moreover, forecasting results clearly demonstrate that significant increases are expected in the future FFCO2 emissions of the BRICS and MINT countries.