基于Logistic模型和FAHP方法的小微客户信用风险管理研究

发布时间：2018-01-13 21:28

本文关键词：基于Logistic模型和FAHP方法的小微客户信用风险管理研究　出处：《南京大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着我国银行业的不断发展,国内商业银行的机构数量和网点数量都在不断增加,国内银行业的竞争环境也在变的日益激烈。到目前为止,对于国内商业银行而言,传统的信贷业务仍然是其主要利润来源。大部分的商业银行都侧重于做好信贷业务,但是单单营销好信贷客户并不能够为其带来稳定的收入,这主要是因为在信贷合同的执行过程中,由于信息的不对称和客户经营风险的动态变化,商业银行不得不承受信贷客户无法归还本金和利息的违约风险。因此为了能够避免信贷客户的信用风险,同时提高在同行业中的竞争力,商业银行必须要建立一套准确可靠的信用风险管理系统。本文依据信用风险管理的相关理论,结合我国商业银行的现实情况,通过选取相关指标,运用Logistic模型和模糊层次分析法,深入探讨了小微客户信用风险同各种财务指标、非财务指标以及小微客户行为事件之间的关系。在模型构建过程中,本文选取了两组小微客户的样本数据,一组数据主要由个体工商户、种养殖户构成,另一组数据主要由小微企业客户构成。在第一组数据中,通过运用Logistic模型进行试验,发现小微客户的很多财务指标,都与其信用风险成显著相关关系；对于第二组数据,由于该数据中的客户仍然持有银行信贷,其信用风险更难以从财务指标中去体现。并且此类客户信用风险形成的主要原因是因为其在经营中遭受到的一些突发事件而导致的,采用小微客户财务报表中的相关指标无法准确判断其目前的信用风险,进而导致了Logistic模型在判断方面的失误。因此本文引入FAHP方法,将小微客户的各种行为事件,包括贷款卡信息、法院诉讼信息等纳入到风险评估体系之中。结果显示在将行为事件纳入到小微客户的信用风险评估体系之后,增强了模型对小微客户信用风险的判断准确性。
[Abstract]:With the continuous development of our banking industry, the number of domestic commercial banks and the number of outlets are increasing, and the competitive environment of domestic banks is also becoming increasingly fierce. So far. For domestic commercial banks, the traditional credit business is still the main source of profits. Most commercial banks focus on doing the credit business well. However, marketing credit customers alone can not bring them stable income, which is mainly due to the asymmetric information and the dynamic changes of customer operating risks in the implementation of credit contracts. Commercial banks have to bear the default risk that the credit customers can not return the principal and interest. Therefore, in order to avoid the credit risk of the credit customers, and at the same time to improve their competitiveness in the same industry. Commercial banks must establish a set of accurate and reliable credit risk management system. According to the relevant theory of credit risk management, combined with the reality of commercial banks in China, through the selection of relevant indicators. Using Logistic model and fuzzy analytic hierarchy process, this paper discusses the credit risk of small and micro customers and various financial indexes. The relationship between non-financial indicators and behavior events of micro-customers. In the process of model construction, this paper selects two sets of sample data of micro-customers, one group of data is mainly composed of individual industrial and commercial households, breed farmers. In the first group of data, by using Logistic model to test, we find a lot of financial indicators of small and micro customers. Both of them have a significant correlation with their credit risk; For the second set of data, since the customer in the data still holds bank credit. Its credit risk is more difficult to reflect from the financial indicators. And the main reason for the formation of this kind of customer credit risk is due to some unexpected events in the operation. It is impossible to judge the credit risk accurately by using the relevant indexes in the financial statements of small and micro customers, which leads to the errors in the judgment of the Logistic model. Therefore, the FAHP method is introduced in this paper. All kinds of behavior events, including credit card information, court litigation information, etc., are incorporated into the risk assessment system. The results show that the behavior events are incorporated into the credit risk assessment system of the small and micro customers. The accuracy of the model to judge the credit risk of small and micro customers is enhanced.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.4;F224

【相似文献】