我国银行间债券市场企业债综合风险评估研究

发布时间：2018-02-27 11:49

本文关键词： 银行间债券市场企业债综合风险评估 Copula函数 VaR值　出处：《湖南大学》2012年硕士论文　论文类型：学位论文

【摘要】：近年来我国银行间债券市场企业债的风险问题逐渐引起社会的广泛关注，如何科学评估企业债综合风险也成为学术界探讨的重要问题。本文以我国银行间债券市场企业债为研宄对象，探宄其综合风险的评估问题。首先，通过分析我国银行间债券市场企业债的综合风险及其评估现状后发现，当前我国企业债实行国家信誉担保，信用风险不突出，而流动性风险和市场风险突显。其次，构建t-GARCH模型以描述流动性风险因子和市场风险因子的边缘分布，，运用Copula函数度量两类风险因子间的相关关系，进而采用蒙特卡洛模拟综合评估银行间债券市场企业债流动性风险和市场风险的整体VaR值。最后，将上述基于Copula函数的蒙特卡洛模拟方法与其他几种传统的风险评估方法相比较，以探宄更为安全、准确的综合风险评估方法。经过理论与实证研宄，本文得到如下结论：基于金融时间序列的尖峰厚尾性、条件方差时变性、波动聚群性-，本文采用的tGARCH模型能够较好地拟合银行间债券市场企业债流动性风险因子与市场风险因子的边缘分布，且两者的波动具有明显的聚群性和持续性；同时Archimedean Copula函数族中的Frank Copula函数可以很好地描述两类风险因子之间的动态非线性耦合关系，且两类风险因子在尾部的相关性加强，呈现对称性；通过实证对比简单加和、联合正态分布和基于Copula函数的蒙特卡洛模拟VaR方法后发现，引入Copula函数后的风险度量更为安全、准确。
[Abstract]:In recent years, the risk of corporate bonds in China's interbank bond market has gradually aroused widespread concern in the society. How to scientifically assess the comprehensive risk of corporate debt has also become an important issue in academic circles. This paper takes the corporate bonds in the interbank bond market of China as the object of study to explore the evaluation of their comprehensive risks. First of all, Based on the analysis of the comprehensive risk of corporate bonds in interbank bond market in China and the present situation of its evaluation, it is found that the credit risk is not prominent, while the liquidity risk and market risk are prominent. Secondly, The t-GARCH model is constructed to describe the edge distribution of liquidity risk factors and market risk factors. The correlation between these two risk factors is measured by Copula function. Then Monte Carlo simulation is used to evaluate the overall VaR value of corporate bond liquidity risk and market risk in the interbank bond market. Finally, The Monte Carlo simulation method based on Copula function is compared with other traditional risk assessment methods in order to explore a more secure and accurate comprehensive risk assessment method. Through theoretical and empirical studies, this paper draws the following conclusions: based on the financial time series, the peak and thick tail, conditional variance time variant, The tGARCH model used in this paper can fit the edge distribution of liquidity risk factor and market risk factor in interbank bond market, and their volatility has obvious clustering and persistence. At the same time, the Frank Copula function in the Archimedean Copula function family can well describe the dynamic nonlinear coupling relationship between the two kinds of risk factors, and the correlation between the two types of risk factors in the tail is strengthened, showing symmetry. After combining normal distribution with Monte Carlo simulation VaR method based on Copula function, it is found that the risk measurement with Copula function is more secure and accurate.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51

【参考文献】