企业碳信贷信用风险的预测模型与应用研究

发布时间：2018-12-10 23:25

【摘要】：近年来,环保问题的关注度持续走高,自2010年两会“一号提案”锁定发展低碳经济以来,我国在不断的探索经济低碳发展、环境可持续发展的实践发展方式。然而,经过几年的实践证明,单纯依靠行政力量的管束或是纯粹的执法手段,我国无法继续走可持续发展道路,更无法完成节能减排的目标。在借鉴国外发达国家低碳经济发展的成功经验的基础上,我国迫切的需要建立一套低碳金融支持低碳经济的管理体系,而处于我国金融行业主体地位的银行,若能积极发展碳信贷业务,这无疑是行之有效的途径之一。本文首先在回顾了社会责任理论、可持续发展理论、国际赤道原则等碳信贷理论基础上,分析了商业银行发展碳信贷的原因。其次,从企业的角度出发,结合碳信贷信用风险的特性,建立企业碳信贷信用风险预测指标体系,选择上市公司作为研究样本,并广泛搜集和整理相关数据,根据2014年起实施的《企业环境信用评价办法》,辅以交通银行2010年8月发布的《交通银行环保标识分类操作手册》中对绿色信贷授信客户环保信息标识的划分细则,并询问专家的意见,评估得到样本公司碳信贷信用风险等级。进而,运用BP神经网络分别从时间序列维度以及截面数据维度建立碳信贷信用风险预测模型,经过网络训练以及仿真等实验操作,对模型预测效果进行可靠性分析并评价两个模型的预测效果。最后,将预测模型应用于实际案例分析,得出结论:时间序列维度较截面数据维度的预测模型更具备实际应用推广价值。
[Abstract]:In recent years, the concern of environmental protection has been increasing continuously. Since the first proposal of the two sessions in 2010 locked in the development of low-carbon economy, China has been exploring the practical development mode of economic low-carbon development and environmental sustainable development. However, after several years of practice, relying solely on administrative control or pure law enforcement means, our country can not continue to take the path of sustainable development, let alone achieve the goal of energy saving and emission reduction. On the basis of the successful experience of the developed countries in the development of low-carbon economy, our country urgently needs to establish a management system of low-carbon finance to support the low-carbon economy, and the banks are in the main position of the financial industry in our country. If can actively develop carbon credit business, this is undoubtedly one of the effective ways. Firstly, based on the review of the theories of social responsibility, sustainable development and international equatorial principle, this paper analyzes the reasons for the development of carbon credit in commercial banks. Secondly, from the point of view of enterprises, combining the characteristics of carbon credit risk, the paper establishes the enterprise carbon credit risk prediction index system, selects listed companies as the research sample, and collects and collates the relevant data widely. According to the measures for Environmental Credit Evaluation of Enterprises implemented since 2014, and supplemented by the detailed rules for the division of green credit customers' environmental information identification, issued by the Bank of Communications in August 2010, the "operational Manual for Classification of Environmental Protection labels of Bank of Communications", And ask expert's opinion, evaluate get sample company carbon credit risk grade. Furthermore, BP neural network is used to establish carbon credit risk prediction model from time series dimension and cross section data dimension respectively, after network training and simulation and other experimental operations. The reliability of the two models is analyzed and the prediction results of the two models are evaluated. Finally, the prediction model is applied to practical case analysis, and it is concluded that the prediction model of time series dimension has more practical application value than that of cross-section data dimension.
【学位授予单位】：南昌大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F832.4;TP183

【参考文献】