基于支持向量机的商业银行绿色信贷评级模型研究

发布时间：2018-10-10 14:20

【摘要】：随着金融自由化、经济全球化和金融创新的发展,商业银行信用风险管理日益迫切与重要。作为社会重要的资金枢纽,商业银行以信贷支持低碳经济的同时面临由环境和社会危机引发的企业信贷风险。绿色信贷的提出为商业银行风险管理提供了新的方向和指标,它能够帮助商业银行评估和管理项目融资过程中所涉及的环境风险和社会风险,同时达到了应用经济手段去实现环境和社会可持续发展的目的。长期以来,我国商业银行信用风险管理体系不健全,没有有效的信用风险模型来控制和衡量企业的违约风险。因此,建立银行信贷风险评估模型和有效的企业信用评估体系是国内银行共同面临的研究课题。现今,随着信息技术的不断发展,人工智能方法和机器学习模型被应用于解决信用评级问题,本文尝试用支持向量机方法对商业银行绿色信贷信用评级做一些探讨和研究。本文首先介绍了信用风险的相关概念,从而引出了信用风险管理；又对国内外常用的信用评级理论和方法进行了介绍和综述,并分析了我国商业银行信用评估及实践绿色信贷的现状与不足,接着概述了支持向量机的理论基础。最后,在理论探讨的基础上进行实证分析,建立了基于绿色信贷的信用评级综合指标体系,构建了基于支持向量机算法的一个信用评估模型,使用Matlab软件为平台实现对企业的二分类。本文采用了理论剖析和实证研究并重的研究方法。将影响商业银行信用风险的贷款企业作为主要的研究对象,然后采用上市公司数据进行了实证分析,应用支持向量机方法建立了全面的商业银行信用评级模型。模型运行后得出如下结论：SVM模型的信用分类能力不错；采用了绿色信贷的信用评级模型的评级分类准确率更优；linear核函数是一种效果比较好的核函数。最后本文对未来的研究及绿色信贷的发展进行了展望与建议。
[Abstract]:With the development of financial liberalization, economic globalization and financial innovation, credit risk management of commercial banks becomes increasingly urgent and important. As an important social capital hub, commercial banks face the credit risk caused by environmental and social crisis while supporting low-carbon economy with credit. Green credit provides a new direction and indicators for the risk management of commercial banks. It can help commercial banks to assess and manage the environmental and social risks involved in the process of project financing. At the same time, the application of economic means to achieve sustainable development of the environment and society. For a long time, the credit risk management system of commercial banks in our country is not perfect, and there is no effective credit risk model to control and measure the default risk of enterprises. Therefore, the establishment of a bank credit risk assessment model and an effective enterprise credit evaluation system are common research topics faced by domestic banks. Nowadays, with the development of information technology, artificial intelligence method and machine learning model are applied to solve the problem of credit rating. This paper attempts to use support vector machine method to do some research on green credit rating of commercial banks. This paper first introduces the related concepts of credit risk, thus leads to the management of credit risk, and then introduces and summarizes the common theories and methods of credit rating at home and abroad. The present situation and deficiency of credit evaluation and practice of green credit of commercial banks in China are analyzed, and then the theoretical basis of support vector machine is summarized. Finally, on the basis of theoretical analysis, a credit rating index system based on green credit is established, and a credit evaluation model based on support vector machine (SVM) algorithm is constructed. Using Matlab software as the platform to achieve the two-classification of enterprises. This paper adopts the research method of both theoretical analysis and empirical research. This paper takes the loan enterprises which affect the credit risk of commercial banks as the main research object, then uses the listed company data to carry on the empirical analysis, uses the support vector machine method to establish the comprehensive commercial bank credit rating model. After the operation of the model, the following conclusions are drawn: the SVM model has good credit classification ability; the credit rating model with green credit has better classification accuracy; and the linear kernel function is a better kernel function. Finally, the future research and the development of green credit are prospected and suggested.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.4

【参考文献】