物流行业客户授信风险研究
发布时间：2018-07-16 23:06
【摘要】：本文将以物流行业的客户主体为研究对象,着手建立区分行业的客户授信风险等级评定指标体系,并且尝试应用一种国际广泛适应的违约风险度量模型——KMV模型来测算预期违约率,完成对客户授信风险等级评定结果精确性检验的返回测试,达到检验建立的客户授信风险等级评定指标体系是否有效的目的。根据股票市场数据度量模型的特点,本文将KMV模型的违约风险度量方法与自身建立的客户授信风险等级评定指标体系的信用评级结果相结合,选取了1301家上市公司作为样本,先采集样本上市公司的财务报表和股票数据,通过运用自编的MATLAB程序计算出代表KMV模型的关键变量——违约距离和预期违约率,再对KMV模型得到的预期违约率和指标体系等级评定结果进行线性回归拟合,形成一一对应的映射关系,最后通过KMV模型计算结果对客户授信风险等级评定指标体系进行样本内检验,结果显示本文自身建立的客户授信风险等级评定指标体系对受评客户违约风险判断的准确率为76.33%。通过本文的研究说明:该指标体系下的授信风险等级对受评客户违约风险具有较强的预测能力,同时KMV模型对上市公司的违约风险考量也具有较大的参考价值。
[Abstract]:This article will take the logistics industry customer main body as the research object, starts to set up the customer credit risk grade appraisal index system which distinguishes the industry, And try to use a kind of international widely adapted default risk measurement model-KMV model to calculate the expected default rate, and complete the return test of the accuracy test of customer credit risk rating results. To verify the effectiveness of the established customer credit risk rating index system. According to the characteristics of the stock market data measurement model, this paper combines the default risk measurement method of the KMV model with the credit rating results of the customer credit risk rating index system established by itself, and selects 1301 listed companies as samples. Firstly, the financial statements and stock data of the listed companies are collected, and the key variables representing the KMV model, the default distance and the expected default rate, are calculated by using the MATLAB program. Then the expected default rate obtained by KMV model and the evaluation results of index system grade are fitted by linear regression to form a one-to-one mapping relationship. Finally, through the KMV model calculation results of customer credit risk rating index system for in-sample test, the results show that the customer credit risk rating evaluation index system established in this paper for the evaluation of customer default risk accuracy is 76.33. Through the research of this paper, it is shown that the credit risk grade under the index system has a strong ability to predict the default risk of the evaluated customer, and the KMV model is also of great reference value to the listed company's default risk evaluation.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F259.2
本文编号：2127968
[Abstract]:This article will take the logistics industry customer main body as the research object, starts to set up the customer credit risk grade appraisal index system which distinguishes the industry, And try to use a kind of international widely adapted default risk measurement model-KMV model to calculate the expected default rate, and complete the return test of the accuracy test of customer credit risk rating results. To verify the effectiveness of the established customer credit risk rating index system. According to the characteristics of the stock market data measurement model, this paper combines the default risk measurement method of the KMV model with the credit rating results of the customer credit risk rating index system established by itself, and selects 1301 listed companies as samples. Firstly, the financial statements and stock data of the listed companies are collected, and the key variables representing the KMV model, the default distance and the expected default rate, are calculated by using the MATLAB program. Then the expected default rate obtained by KMV model and the evaluation results of index system grade are fitted by linear regression to form a one-to-one mapping relationship. Finally, through the KMV model calculation results of customer credit risk rating index system for in-sample test, the results show that the customer credit risk rating evaluation index system established in this paper for the evaluation of customer default risk accuracy is 76.33. Through the research of this paper, it is shown that the credit risk grade under the index system has a strong ability to predict the default risk of the evaluated customer, and the KMV model is also of great reference value to the listed company's default risk evaluation.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F259.2
【参考文献】
相关期刊论文 前1条
1 窦文章;刘西;;基于CreditMetrics模型评估银行信贷的信用风险[J];改革与战略;2008年10期
，本文编号：2127968
本文链接：https://www.wllwen.com/jingjilunwen/jingjiguanlilunwen/2127968.html
