Research on Credit Risk Evaluation of Green Credit Based on BP Neural Network
|School||Ocean University of China|
|Keywords||Green Credit Credit risk evaluation BP neural network Gountermeasures and suggestions|
In the21st century, human society’s resources and environment problem increasingly highlighted. It has become a consensus to adjust the economic structure, transform the mode of economic growth and develop the low carbon sustainable. While capital is as the economic development of the blood, and bank is as capital allocation hub, combining with the social responsibility requirements of banks, make banks play a vital role in these. Adjust the structure of bank credit idea and development, concerned about the environment risk, low carbon finance has become the international banking act. At present, most of the major international financial institutions have joined the "equator principles", and the carbon financial theories and practices have get fast development. Our country has put forward the Green Credit in2007, under the pressure of internal environment and economic transition and the international carbon finance development trend, to adjust social capital flows so as to cut off the capital for "two high one left" enterprises’development to support the development of low-carbon eco-friendly industries and enterprises. At the same time, however, the Green Credit makes bank face a complex and changing environmental risks. And the deficiencies of risk assessment and management of Green Credit has hindered the development of it. Therefore, risk evaluation and management mechanism of Green Credit must be established, and the risk evaluation index system and evaluation model of it should be explored and perfected, to provide a theoretical basis for the smooth development of green credit carry.This article is from the perspective of the credit risk evaluation, aims to build the credit risk evaluation index system of Green Credit basing on carding of theories about Green Credit and its credit risk evaluation. And then it uses the BP neural network with a low carbon environmental protection of listed companies as samples for empirical analysis. Specifically, this paper includes six parts as follows:In part I, related theories like sustainable finance theory, the bank of environmental risk management theory, theory of corporate social responsibility, and bank credit risk management theory are introduced as the theoretical basis. Part II is about qualitative study of Green Credit’s credit risk evaluation. Manifestation of the Green Credit’s credit risk characteristics and causes are analyzed based on the introduction of green credit connotation and environmental risks. After then, it elaborates the connotation, present situation and problems about its credit risk evaluation. Part Ⅲconstructs credit risk evaluation index system and model of Green Credit. On the basis of the existing the green credit risk evaluation index system and its characteristics, the paper built an index system containing traditional financial indicators and environmental risk indicators. Among them, environmental risk indicators include social contribution value per share and other four corporate performances, as well as the environmental quality like the "three wastes" emissions. Then, the credit risk evaluation model of Green Credit is constructed based on the principle of BP neural network. Part Ⅳ is case analysis. Selecting the listed companies of110low-carbon environmental as samples, and then use of the index system and model to network train, test and use. And the credit risk profile of Green Creditis evaluated according to the results. In part Ⅴ,some countermeasures and suggestions about credit risk management of Green Credit are put forward from aspects of external institutional environment and internal control. Finally, the article makes a summary and an outlook from the research method sand ideasThe innovation of this paper is as follow:(1) it from the perspective of corporate credit risk of Green Credit. It chooses the low-carbon environment of listed companies as samples to make risk evaluation, which provides a new idea for Green Credit’s credit risk evaluation and management.(2) At empirical aspect, it introduces social contribution value per share, the energy consumption per unit of output, the "three wastes" emissions and other seven environmental risk indicators based on Green Credit’s credit risk characteristics and data available, to build risk evaluation index system of Green Credit. And then it uses the BP neural network to make quantitative case analysis, which makes exploratory research for the credit risk evaluation model and method of Green Credit.