Application of Carbon Emissions Prediction Using Least Squares Support Vector Machine Based on Grid Search

Authors: , ,

Abstract: The rapid development of the global economy has brought a high-speed increase in carbon emissions, which is the primary cause of the greenhouse effect, thus how to effectively alleviate the economic and ecological threats caused by greenhouse effect, as well as studies about carbon emissions mechanism and related fields are all of great urgency. This paper puts forward a new method for carbon emissions prediction--least squares support vector machine based on grid search, where grid search is used to optimize the regularization parameter and kernel width parameter in least squares support vector machine. Then choose population, per capita GDP, energy consumption per unit GDP, urbanization rate, the proportion of coal consumption and proportion of value added services these six influencing factors as independent variables, for the aim to predict China’s annual carbon emissions. The results show that the mean absolute percentage error of this new method is 1.61%, superior to that of BP neural network model, illustrating the proposed can be effectively used to predict the carbon emissions.

WSEAS Transactions on Systems and Control, ISSN / E-ISSN: 1991-8763 / 2224-2856, Volume 10, 2015, Art. #11

PDF
XML
Cite
×
Citation Tools
WSEAS AMA
Wei Sun, Yi Liang, Yanfeng Xu, "Application of Carbon Emissions Prediction Using Least Squares Support Vector Machine Based on Grid Search," WSEAS Transactions on Systems and Control, vol. 10, pp. 95-104, 2015, DOI:
Wei Sun, Yi Liang, Yanfeng Xu. Application of Carbon Emissions Prediction Using Least Squares Support Vector Machine Based on Grid Search. WSEAS Transactions on Systems and Control. 2015;10:95-104.
Copy to Clipboard
Citation copied to Clipboard

Certification

---