工商银行湖南分行内部信用评级管理研究

发布时间：2018-11-02 08:54

【摘要】：中国的商业银行开展客户评级、贷款评级的的时间不是太长，数量也不多，目前还没有一套科学的成熟可靠的信用评级体系。各以工农中建四大国有商业银行为首的中国商业银行正在摸索和建设自己的内部评级体系。但是我国目前既没有建立一个比较权威性的第三方评估机构，又缺乏一套科学而有效的评级体系。没有制定行业内通行的统一规范，各家商业银行推行自己的客户评级标准。同一公司客户在不同银行的信用评级结果差异颇大。而国外发达国家，既有正规的第三方评级机构，也有可靠而有效的评级体系。国内商业银行的信用风险管理水平和意识还有待加强和提高。由于信用评级结果的可靠性取决于参与评级客户所提供数据的真实性，故根据银行内部信贷数据来构建一套客户信用风险评级体系显得尤为重要。本文首先介绍了信用风险的定义、新巴塞尔协议下的内部评级法（IRB）的主要内容。对比分析了国内外对违约概率模型PD的研究情况。接着介绍了工商银行湖南分行内部信用评级工作的开展情况，对比分析了工商银行内部评级一期和二期项目所取得的成就及不足之处。最后运用Logistic回归模型对湖南工行所属的企业客户进行实证分析，构建湖南工行企业信用风险评估模型。在模型的构建时，随机选取了72家企业作为建模样本组，随机选取了58家企业作为检验样本组，通过处理样本企业2012年的数据得到相应的40个财务指标，利用SPSS软件进行正态性检验、假设检验、相关性检验来筛选指标，最后再利用Logistic回归模型对筛选出来的指标进行回归分析。通过使用检验样本组对该模型进行验证，发现该模型的准确率达到了84.5%。本文所得出的公司客户信用评价模型对于湖南工行现有的情况来说是较为有效的。
[Abstract]:China's commercial banks carry out customer rating, loan rating is not too long, the number is not much, there is no scientific and reliable credit rating system. China's commercial banks, led by four state-owned commercial banks, are groping and building their own internal rating system. However, there is neither an authoritative third-party evaluation institution nor a scientific and effective rating system in China. There is no uniform standard in the industry, commercial banks promote their own customer rating standards. The credit rating results of the same company's customers in different banks vary greatly. Foreign developed countries, both formal third-party rating agencies, but also reliable and effective rating system. The credit risk management level and consciousness of domestic commercial banks need to be strengthened and improved. Because the reliability of the credit rating results depends on the authenticity of the data provided by the participating rating customers, it is particularly important to construct a set of customer credit risk rating system based on the internal credit data of the bank. This paper first introduces the definition of credit risk and the main content of (IRB). The research of default probability model (PD) at home and abroad is compared and analyzed. Then it introduces the development of the internal credit rating of ICBC Hunan Branch, and compares and analyzes the achievements and shortcomings of ICBC's internal rating projects in the first and second phases. Finally, the Logistic regression model is used to analyze the customers of Hunan Industrial and Commercial Bank of China (ICBC), and the credit risk assessment model of ICBC is constructed. When the model was constructed, 72 enterprises were randomly selected as modeling sample group, 58 enterprises were randomly selected as test sample group, and 40 financial indexes were obtained by processing the data of sample enterprises in 2012. The SPSS software is used for normality test, hypothesis test and correlation test to screen the indexes. Finally, the Logistic regression model is used to analyze the selected indexes. It is found that the accuracy of the model is 84.55by using the test sample group to verify the model. The customer credit evaluation model obtained in this paper is effective for the current situation of Hunan Industrial and Commercial Bank of China (ICBC).
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.33

【参考文献】