中国商业银行信用风险度量实证研究

发布时间：2018-04-26 00:13

本文选题：信用风险 + KMV模型　；参考：《江西财经大学》2012年硕士论文

【摘要】：信用风险的度量是商业银行经营的一个永恒话题。近年来,随着金融全球化和经济全球化进程的加快,商业银行面临的信用风险越来越大,亦越来越复杂。因此,如何有效地控制和度量信用风险成为社会各界人士关注的焦点。由于中国经济体制的特殊性,目前中国商业银行主要还是使用传统的方法(如信用分析方法、专家经验判断法等)对信用风险进行度量,这样的度量方式远不能满足商业银行对企业信用风险度量的要求。因此,研究国外先进的信用风险度量模型,开发适合中国国情的信用风险度量模型,对提高中国银行业信用风险管理的能力具有重要意义。本文从信用风险的界定、风险波动性、风险驱动因素和相关性、回收率以及适用范围等五个方面分析比较现代信用风险度量方法在中国的适用性。根据目前中国商业银行的信用风险特征和管理现状,选取KMV模型,并对模型中的股权和违约点的确定做了一定的修正,以中国上市公司的数据为样本,对中国商业银行信用风险度量进行实证分析,得到的主要结论如下： (1)利用违约距离和违约概率度量上市公司的违约情况,效果良好。由于目前中国还没有像KMV公司一样拥有庞大的数据库,因此无法通过违约距离得到经验EDF,但利用违约距离和违约概率度量上市公司的违约情况,也起到了很好的效果。实证中显示,ST公司的信用风险一般都比正常公司的信用风险大。表明此方法对中国上市公司的信用风险有一定的预测能力,从而警示银行及早作好防范决策。 (2)把违约点设定为“流动负债+0.75×长期负债”,得到的结果更适合中国企业的信用状况。在KMV模型中,违约点的确定是该模型的主要步骤之一。KMV公司通过大量违约事件进行验证,当违约点为“流动负债+0.5×长期负债”时,最能反映上市公司的违约情况。对于这个违约点的适用性是否符合中国的经济体制,这个值得我们考虑。因此,本文在选用KMV公司确定的违约点基础上,增加一个违约点,即取“流动负债+0.75×长期负债”,并分别对这两个违约点计算出它们各自的违约距离和违约概率,然后通过简单的均值检验以及T检验方法判定这两种违约点的适用性。两种违约点均能通过检验,但是当“违约点=流动负债+0.75×长期负债”时,所得到的结果更适合中国基本情况。这与KMV公司所确定的违约点有些差异,导致这种差异的主要原因是由于中国上市企业信用的严重缺失。从客观上来说,只有当企业的负债总额高于企业资产价值的一定比例时,企业才会出现违约现象。因此,KMV公司将违约点设在“流动负债+0.5×长期负债”是比较合适的。但由于中国上市企业的信用缺失情形比较严重,很多企业为避免亏损,当其资产价值出现下滑但还未低于负债总额时,就会出现违约情况。因此,把违约点设定为“流动负债+0.75×长期负债”是与中国企业信用状况相适应的。加强中国商业银行信用风险管理,提高中国商业银行信用风险度量的准确性、科学性,一要加强现代信用风险管理文化意识；二要建立健全的相关信用机制；三要建立信用数据库；四要建立科学有效的信用风险度量模型,对信用风险实现全方位的度量。
[Abstract]:The measurement of credit risk is an eternal topic for commercial banks. In recent years, with the rapid development of financial globalization and economic globalization, the credit risk of commercial banks is becoming more and more complex. Therefore, how to effectively control and measure credit risk has become the focus of attention of all circles of society. At present, the commercial banks of China mainly use the traditional methods (such as credit analysis method, expert experience judgment method, etc.) to measure the credit risk. This measure can not meet the requirements of the commercial bank's credit risk measurement. The credit risk measurement model suitable for China's national conditions is of great significance for improving the credit risk management ability of China's banking industry.
This paper compares the applicability of the modern credit risk measurement method in China from five aspects, such as the definition of credit risk, risk volatility, risk driving factors and relevance, recovery rate and the scope of application. According to the current credit risk characteristics and management status of China's commercial banks, the KMV model is selected and the equity and violation in the model are made and violated. With the data of Chinese listed companies as samples, the empirical analysis of the credit risk measurement of Chinese commercial banks is carried out. The main conclusions are as follows:
(1) the default distance and default probability are used to measure the default situation of the listed companies, and the effect is good. Since there is no huge database like KMV company in China at present, it can not get experience EDF through default distance, but it has also played a very good effect by measuring default distance and default probability on the default situation of the company. The demonstration shows that the credit risk of ST company is generally greater than that of the normal company. It shows that this method has a certain predictive ability for the credit risk of Chinese listed companies, thus warning the bank to make early preventive decisions.
(2) the default point is set as "+0.75 * long-term liabilities of current liabilities". The results obtained are more suitable for the credit status of Chinese enterprises. In the KMV model, the determination of the default point is one of the main steps of the model..KMV company is verified by a large number of default events. When the default point is "current liabilities +0.5 x long term liabilities", it can best reflect It is worth considering whether the applicability of the city is in conformity with the economic system of China, which is worth considering. Therefore, on the basis of the default point determined by KMV company, this paper adds a default point, that is, the "current liabilities +0.75 x long liabilities", and to calculate their respective violation of the two default points respectively. About distance and default probability, and then using a simple mean test and T test to determine the applicability of the two default points. Two kinds of default points can be tested, but when the "default = +0.75 x long liabilities", the results are more suitable for China's basic situation. This is somewhat inferior to the default point identified by KMV company. The main reason for this difference is due to the serious lack of credit in China's listed companies. Objectively, only when the total amount of liabilities of the enterprise is higher than the value of the enterprise asset value, the enterprise will appear to be in breach of contract. Therefore, it is more appropriate for KMV company to set a default point in the "+0.5 x long liability" of the "current liabilities". However, due to the serious lack of credit in China's listed companies, many enterprises in order to avoid losses, when the value of assets decline but is still not lower than the total liabilities, there will be a default situation. Therefore, the default point is set as "+0.75 * long-term liabilities of current liabilities" is compatible with the credit status of Chinese enterprises.
To strengthen the credit risk management of Chinese commercial banks and to improve the accuracy and scientificity of the credit risk measurement of China's commercial banks, we should strengthen the cultural awareness of modern credit risk management; two to establish a sound related credit mechanism; three to establish a credit database; four to establish a scientific and effective credit risk measurement model, and to the credit risk. Achieve an omni-directional measure.

【学位授予单位】：江西财经大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.33;F224

【参考文献】