Research On The Green Credit Risk Assessment Model Of Listed Companies Based On RF And Integrated SVM

The rapid growth of China’s economy has led to a sharp rise in energy consumption.Natural environmental problems such as serious environmental pollution and imbalanced energy structure have become more and more prominent.In order to optimize the adjustment of energy structure,promote the rapid development of green environmental protection industry,and protect the ecological environment,the state has continuously emphasized the importance of civilized construction and the establishment of a green financial development system.From the perspective of the development of green finance in China,green credit is the main way for commercial banks in China to practice green finance.Since the launch of Green Credit in 2007,it has received full support from the government and banks.However,in order to achieve the goal of controlling the development of high-pollution and high-energy industries,green credit has more diverse risks than traditional credit,and its risk assessment is more complicated.The particularity of green credit makes the conventional credit risk assessment method difficult to apply.Therefore,it is very important to establish a comprehensive and objective green credit risk assessment indicator system and model to conduct reasonable and effective risk assessment.In this paper,2,906 listed companies were selected as research objects.The index system including 18 indicators of financial indicators,technical indicators and environmental indicators.The random forest model was used to sort the importance of the indicators,and finally the group retaining 15 indicators was obtained.Then the SVM is used to establish the model and the SVM established by Gaussian kernel function hasexcellent performance in green credit risk assessment.Then,the SVM was optimized by using ensemble learning algorithm.