Prediction and analysis of the three major industries and residential consumption CO2 emissions based on least squares support vector machine in China

Abstract

In recent decades, the surge in China's CO2 emissions has caused serious environmental problem. According to the CO2 reduction commitment in 12th Five-Year Plan, this paper classifies total CO2 emissions in accordance with industry and analyses CO2 emissions changing trend of residential consumption and the three major industries which include the primary, secondary and tertiary industry. Cointegration and Granger causality test are proposed to select different influencing factors of three major industries and residential consumption CO2 emissions respectively as well as check the influencing factors with different leading lengths. Least squares support vector machine (LSSVM) is applied to predict different types of carbon dioxide emissions. Case studies reveal that classification and prediction of CO2 emissions can highly improve forecast accuracy. The effectiveness of LSSVM has been proved by the final simulation which shows that the proposed method outperforms logistic model, back propagation artificial neural network (BPNN) and GM (1, 1) model in the CO2 emissions forecasting.