基于CPV模型的我国商业银行信用风险度量

发布时间：2018-11-13 14:46

【摘要】：在2008年全球金融危机中,美国国内多家银行遭遇了倒闭的厄运。信用风险的管理缺陷为这次金融危机埋下了种子。目前,我国商业银行中间业务占比较小,主要以赚取净息差收入为主要盈利来源,贷款资金在总资产中占有很大的比重。虽然我国的不良贷款率曾大幅度下降,近些年来一直保持在较低的水平。但这并非表示我国商业银行对于风险管理水平较高。由于历史因素,我国的不良贷款的结构存在很大的问题,风险管理水平较西方发达国家相比也存在很大差距。且近几年,不良贷款率有不断上升的趋势。对于我国商业银行来说,加强信用风险的管理刻不容缓。强化信用风险的管理首先要以恰当的方式来度量风险,才能有的放矢,制定对应的风险管理对策。这是本文研究的主旨所在。在信用风险度量的研究方面,定性的风险测量方式在早期的风险测算中占据主体地位,近些年来更多地采用定量分析方式,较多采用的是现代风险度量模型。CPV模型是其中的一种,该模型以一个国家或地区的宏观经济因素为测算依据,充分考虑到了宏观经济因素对于商业银行信用风险的影响。不仅可以对信用风险进行测算,还可以找出影响信用风险的因素以及各因素对于信用风险的影响程度。这对于风险管控部门预测风险、预防风险提供了充分的决策依据。就目前的我国商业银行来讲,以宏观经济因素为依据的CPV风险度量方式更加适用于我国资本市场不完善的金融大环境。现代的信用风险度量模型分为四种。在对四种模型进行简要分析之后,我们可以发现各个模型各具优缺点,也具有不同的适用性。KMV模型在资本市场以及信用管理水平较高的地区更具适用性；Credit Risk+模型较适用于贷款组合的风险度量；Credit Metrics模型对于数据的要求较高；而CPV模型可以有效地解决上述模型存在的这些问题,具有数据易于获取、考虑全面、准确性强等特点。就这四个模型而言,CPV模型更加适用于我国商业银行信用风险的测度。在实证研究部分,本文首先简要说明CPV模型的原理以及建模步骤。随后,选用了相关宏观经济指标利用CPV模型进行实证分析。根据全面性、代表性、易得性的原则同时参考了前人的研究经验选取了七个宏观经济指标。分别为国内生产总值(GDP)、消费者价格指数(CPI)、城镇居民人均可支配收入(SR)、固定资产投资总额(GD)、社会消费品零售总额(SXL)、狭义货币供应量(M1)、财政支出总额(CZ)。数据均来源于中国统计年鉴公布的2005年第一季度到2015年第三季度的季度数据。其中,2005年第一季度至2015年第二季度组成的样本作为建模样本,2015年第三季度的样本作为检验样本。然后,进行了指标筛选与数据的预处理。采用SPSS的逐步进入的方式进行了指标筛选。利用CPI指数法消除了通货膨胀因素,利用十二步移动平均法消除了季节因素,利用指标的对数化处理消除了异方差。通过所得模型发现,财政支出总额(CZ)、狭义货币供应量(M1)与我国商业银行的不良贷款率呈现负向相关关系,固定资产投资总额(GD)、消费者价格指数(CPI)、城镇居民人均可支配收入(SR)与不良贷款率呈现正向相关关系。
[Abstract]:In the global financial crisis of 2008, many of the banks in the United States have suffered from failure. The management of credit risk buried the seed in the financial crisis. At present, the middle business of the commercial bank of our country is relatively small, mainly to earn net interest income as the main profit source, the loan fund has a large proportion in the total assets. Although the rate of non-performing loans in our country has declined substantially, it has been at a lower level in recent years. But it does not mean that China's commercial banks are relatively high in risk management. Due to the historical factors, the structure of the non-performing loans in China has a big problem, and the risk management level also has a great gap compared with the western developed countries. In recent years, the rate of non-performing loans has a rising trend. It is urgent to strengthen the management of credit risk for commercial banks of our country. To strengthen the management of credit risk, the risk can be measured in an appropriate way, and the corresponding risk management countermeasures can be set up. This is the main subject of this study. In the aspect of the research of the credit risk measure, the qualitative risk measurement method takes the status of the main body in the early risk measurement, and the quantitative analysis method is adopted in recent years, and the modern risk measurement model is more adopted in recent years. The CPV model is one of them, which is based on the macro-economic factors of a country or region, and takes fully into account the impact of the macro-economic factors on the credit risk of commercial banks. Not only can the credit risk be measured, but also the factors that affect the credit risk and the degree of influence of each factor on the credit risk can be found. This provides a sufficient basis for decision-making for risk control, risk prevention and risk prevention. In terms of the current Chinese commercial banks, the measure of CPV risk based on the macro-economic factors is more applicable to the imperfect financial environment of our country's capital market. The modern credit risk measurement model is divided into four categories. After a brief analysis of the four models, we can find the advantages and disadvantages of each model, and also have different applicability. The KMV model is more applicable in the capital market and the higher credit management level; the Credit Risk + model is more applicable to the risk measure of the loan combination; the Credit Metrics model is higher for data; and the CPV model can effectively solve the problems existing in the model. The method has the characteristics of easy acquisition of data, comprehensive consideration, strong accuracy and the like. For these four models, the CPV model is more suitable for the measurement of the credit risk of commercial banks in China. In the part of the empirical research, this paper first briefly describes the principle of CPV model and the modeling steps. Then, the relevant macroeconomic indicators were selected to use the CPV model to carry out the empirical analysis. According to the principle of comprehensiveness, representativeness and availability, seven macroeconomic indicators have been selected by reference to the previous experience. It is the gross domestic product (GDP), the consumer price index (CPI), the per capita disposable income (SR) of the urban residents, the total investment of fixed assets (GD), the total retail sales of the social consumer goods (SXL), the narrow money supply (M1) and the total expenditure (CZ). The data is derived from the quarterly data from the first quarter of 2005 to the third quarter of 2015 published by the China Statistical Yearbook. The samples made up from the first quarter of 2005 to the second quarter of 2015 were used as the sample of construction, and the samples in the third quarter of 2015 were used as test samples. then, the pre-processing of the index screening and the data is carried out. The index selection was carried out by the step-by-step approach of SPSS. The factors of inflation are eliminated by the CPI index method, the seasonal factors are eliminated by means of the 12-step moving average method, and the variance is eliminated by the logarithmic processing of the index. Through the obtained model, the total expenditure (CZ), the narrow money supply (M1) and the non-performing loan ratio of the commercial banks in China are negatively related, the total investment (GD) and the consumer price index (CPI) of the fixed assets, The per capita disposable income (SR) of urban residents is positively related to the rate of non-performing loans.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：F832.33

【相似文献】