中国系统性金融风险的识别、预警与防范研究

发布时间：2018-01-21 04:26

本文关键词： 系统性金融风险金融压力指数极值理论 Logit模型 KLR信号法宏观审慎监管　出处：《华中科技大学》2015年硕士论文　论文类型：学位论文

【摘要】：完善的金融市场是一个发达国家的具体体现,资金资源的合理转移和分配能有效提高实体经济的运转效率。但依托资金与信用的金融体系有其固有的风险,金融市场一旦崩溃,对实体经济将产生不可估量的损失。由于严格的监控和管制,我国目前为止没有爆发过大规模的系统性金融危机,但随着国际金融市场的联动性日益增强,国内金融市场的改革逐步深化,金融创新日新月异,针对我国系统性金融风险的识别、预警和防范进行研究具备十分重要的意义。本文以系统性金融风险的识别、预警与防范为核心研究对象,首先介绍了系统性金融风险的基本理论,然后归纳整理了国内外学者的研究成果,重点在于对系统性金融风险的测度与识别,预警与防范。实证研究中,本文从银行业压力、资产泡沫、外汇市场压力三个维度构建中国金融压力指数FSI,并基于极值理论、广义帕累托分布对我国的金融风险期进行了识别。以FSI同期二元风险变量为基础构造因变量,从国内宏观经济层面、国内金融系统层面、经常账户-国际贸易、资本账户-国际金融四个层面选取了21个变量作为系统性金融风险的备选预警指标,通过ADF平稳性检验、Granger因果关系检验、因子分析法从备选预警指标体系中筛选出平稳的,与FSI具有格兰杰因果关系的变量,并最终提取出5个公共因子作为自变量,运用Logit模型预测了系统性金融风险概率,进一步验证了Logit模型样本内、样本外的预测效果。此外,本文还利用KLR信号法确定了危机发生的概率阈值。最后,本文从宏观审慎监管的角度论述了我国对系统性金融风险的防范工作。本文研究结果表明:FSI指数能较好地拟合我国金融市场的压力状况,2003年、2007年、2008年、2009年、2010年是我国金融市场压力较大,风险爆发可能性较大的年份。因子分析法显示宏观实体经济、外汇贸易、利率变化、通货膨胀和资产市场泡沫是我国金融市场压力的主要来源。根据KLR噪音信号比法确定的风险概率阈值是34%,Logit模型预备较好的样本外预测能力。从宏观审慎监管的角度来看,加强央行、银监局、保监局、证监局的协同监管作用;区分监管与扶持的分界点,深化存款保险制度;信息透明化,加大对金融产品风险收益的信息披露;加强对影子银行的风险监管是防范金融危机的必要举措。
[Abstract]:The perfect financial market is the concrete embodiment of a developed country, the rational transfer and distribution of capital resources can effectively improve the efficiency of the real economy, but the financial system relying on funds and credit has its inherent risks. Once the financial market collapses, there will be inestimable losses to the real economy. Due to strict monitoring and control, China has not broken out a large-scale systemic financial crisis so far. However, with the increasing interaction of international financial markets, the reform of domestic financial markets is gradually deepening, and financial innovation is changing with each passing day, aiming at the identification of systemic financial risks in China. The research on early warning and prevention is of great significance. This paper focuses on the identification, early warning and prevention of systemic financial risk. Firstly, it introduces the basic theory of systemic financial risk. Then summarized the domestic and foreign scholars' research results, focusing on the measurement and identification of systemic financial risks, early warning and prevention. Empirical research, this paper from the banking pressure, asset bubbles. The foreign exchange market pressure three dimensions construct the Chinese financial pressure index FSIs, and based on the extreme value theory. The generalized Pareto distribution identifies the financial risk period in China. Based on the dual risk variables of FSI, the structural dependent variables are from the domestic macroeconomic level and the domestic financial system level. Current account-international trade, capital account-international finance four levels of the selection of 21 variables as an alternative early warning indicators of systemic financial risk, through the ADF smoothness test. Granger causality test, factor analysis from the alternative early warning index system to screen out the stable and FSI with Granger causality variables. Finally, five common factors are extracted as independent variables, and the probability of systemic financial risk is predicted by using Logit model, which further verifies the prediction effect in and outside the sample of Logit model. This paper also uses the KLR signal method to determine the probability threshold of the crisis. Finally. This paper discusses the prevention of systemic financial risks in China from the perspective of macro-prudential supervision. The research results show that the ratio FSI index can better fit the pressure situation of China's financial market, 2003. 2007, 2008, 2009, 2009 is the year that our country financial market pressure is big, risk burst possibility is bigger. Factor analysis shows macroscopical real economy, foreign exchange trade. Interest rate change, inflation and asset market bubble are the main sources of financial market pressure in China. The risk probability threshold determined by KLR noise signal ratio method is 34%. From the point of view of macro-prudential supervision, strengthen the cooperative supervision role of the central bank, the bank supervision bureau, the insurance bureau and the securities bureau; Distinguish between supervision and support, deepen deposit insurance system; Transparency of information to increase the disclosure of financial product risk returns; Strengthening the risk supervision of shadow banks is a necessary measure to prevent financial crisis.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F832

【参考文献】