上市公司信用风险度量模型的探讨

发布时间：2018-06-22 15:25

本文选题：信用风险 + 判别分析模型　；参考：《西南财经大学》2014年硕士论文

【摘要】：商业银行所面临的信用风险是银行业所不容忽视,能否有效地度量信用风险更是影响着银行业的长足发展。随着我国资本市场的不断发展,各个行业的风险不断增加,然而正处于转轨与新兴发展阶段的我国商业银行在信用风险的管理和度量方面的研究相对落后。这就需要我国商业银行借鉴国际上发展较为成熟的银行的风险度量模式,根据我国银行业的经营特点和风险环境,建立合适的信用分析度量模式。本文选择上市公司的信用风险度量为研究方向,试图在梳理国内外研究成果的基础上,对公司信用风险度量进行较为系统的理论分析,并选取样本进行实证分析,对所选择的信用风险度量模型的适用性进行一些探索性分析。本文从信用风险的度量方法出发,详细介绍了当前信用风险度量中的定性和定量分析方法,并结合我国商业银行的现状,分别采用判别分析模型与Logistic模型对所选取的ST公司与非ST公司进行分析,建立了各自的分类依据,从而考察两个模型的判别准确率；接着对KMV模型的内容和实施步骤进行详尽的说明,在此基础上,选取了15家ST公司与配对的15家非ST公司,分别计算了样本公司在不同违约点下的违约距离,对ST公司与非ST公司的违约距离进行Wilcoxon秩和检验,并对本文所采用的三种模型进行对比,考察KMV模型的信用风险识别能力和在我国上市公司风险度量中的适用性；最后对实证分析结果进行总结,指出本文存在的不足,并就当前我国信用风险管理现状提出相应的建议。本文的内容分为五个章节：第一章为绪论这一部分主要介绍文章的选题背景、相关文献以及本文内容安排等。第二章为信用风险与信用风险度量方法这一章节主要对信用风险的相关概念进行解释,较为详细地介绍了不同信用风险度量模型的内容,并根据各个模型的特点选出本文所要采用的信用风险度量模型。第三章为基于多元判别分析模型与Logistic模型的上市公司信用风险度量本章节通过选取合适的样本公司和财务指标,分别采用多元判别分析模型与Logistic模型对样本数据进行分析和比较。第四章为基于KMV模型的信用风险度量基于对KMV模型的详细介绍,选取满足一定标准的样本公司,对模型的参数进行设定,分别求解出上市公司资产价值、波动率、违约点、违约距离,并检验两类公司的违约距离是否有显著差异,最后通过ROC曲线将本文所采用的三种度量模型进行对比。第五章为结论和建议对本文的研究结果和不足进行说明,并就我国信用风险的管控现状提出几点建议。
[Abstract]:The credit risk faced by commercial banks can not be ignored by the banking industry, and whether the credit risk can be effectively measured affects the rapid development of the banking industry. With the development of the capital market in China, the risks of various industries are increasing. However, the research on credit risk management and measurement of commercial banks in China is relatively backward in the stage of transition and emerging development. This requires our commercial banks to draw lessons from the more mature international banks risk measurement model, according to the operating characteristics and risk environment of our banking industry, to establish a suitable credit analysis measurement model. This article chooses the listed company's credit risk measurement as the research direction, tries to comb the domestic and foreign research results, carries on the relatively systematic theory analysis to the company credit risk measurement, and selects the sample to carry on the empirical analysis. The applicability of the selected credit risk measurement model is analyzed. Starting from the measurement method of credit risk, this paper introduces the qualitative and quantitative methods of credit risk measurement in detail, and combines the present situation of commercial banks in our country. The discriminant analysis model and Logistic model are used to analyze the selected St company and non-St company respectively, and the classification basis is established to investigate the discriminant accuracy of the two models. Then, the contents and implementation steps of KMV model are explained in detail. On this basis, 15 St companies and 15 matched non-St companies are selected to calculate the default distance of the sample companies under different default points. The Wilcoxon rank sum test of the default distance between St company and non-St company is carried out, and the three models used in this paper are compared to investigate the credit risk identification ability of KMV model and its applicability in the risk measurement of listed companies in China. Finally, it summarizes the results of empirical analysis, points out the shortcomings of this paper, and puts forward corresponding suggestions on the current situation of credit risk management in China. The content of this paper is divided into five chapters: the first chapter is the introduction of this part mainly introduces the background of the article, relevant literature and the content of this article. The second chapter mainly explains the related concepts of credit risk and the measurement methods of credit risk, and introduces the content of different credit risk measurement models in detail. According to the characteristics of each model, this paper chooses the credit risk measurement model. The third chapter is the credit risk measurement of listed companies based on multivariate discriminant analysis model and Logistic model. This chapter selects appropriate sample companies and financial indicators. Multivariate discriminant analysis model and Logistic model were used to analyze and compare the sample data. The fourth chapter is the credit risk measurement based on KMV model. Based on the detailed introduction of KMV model, select the sample companies that meet certain standards, set the parameters of the model, calculate the asset value, volatility, default point of listed companies, respectively. The distance of default and the difference of default distance between the two types of companies are tested. Finally, the three measurement models used in this paper are compared by ROC curve. The fifth chapter is the conclusion and the suggestion to explain the research result and the insufficiency of this paper, and puts forward some suggestions on the present situation of the credit risk control in our country.
【学位授予单位】：西南财经大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51;F203;F224

【参考文献】