两因子CIR对上交所国债利率期限结构的实证研究

发布时间：2018-06-09 16:54

本文选题：利率期限结构 + 两因子CIR　；参考：《安徽财经大学》2012年硕士论文

【摘要】：随着利率市场化和国债市场规模的不断壮大,隐含在国债中的利率期限结构有着重要的意义,它是资产定价、风险管理和套期保值的基础,也是中央银行制定货币政策的分析工具。因此,研究利率期限结构有着重要的意义。利率期限结构理论分为传统利率期限结构理论和现代的静态理论以及动态理论。本文首先分析了国债利率期限结构的研究背景和意义以及国内外关于利率期限结构研究情况,在论文第二章中系统地介绍传统利率期限结构理论和利率期限结构模型理论。第三章详细阐述两因子CIR模型的参数估计方法即卡尔曼滤波法。本文的实证研究部分,首先运用Nelson-Siegel模型产生即期利率的数据。通过对各期利率相关系数的分析,得出单因子模型不能够准确描述利率的动态变化过程。而因子分析说明两因子模型已经能够描述利率的绝大部分动态变化特征。其次在即期利率的基础上,通过卡尔曼滤波法得出两因子CIR模型的参数。最后由两因子CIR模型估计的误差均方根数值可以看出,对于1年期即期利率模型的拟合不够精确,但能够较为准确刻画其他期限较长的即期利率数据。在整体上,根据卡尔曼滤波方法所得出的数据和实际数据较为相近。因而,两因子CIR模型可以比较准确地刻画我国国债市场利率期限结构的动态变化过程。
[Abstract]:With the marketization of interest rate and the growing scale of the national debt market, the term structure of interest rate implied in the national debt is of great significance. It is the basis of asset pricing, risk management and hedging. It is also an analytical tool for central banks to set monetary policy. Therefore, it is of great significance to study the term structure of interest rate. The term structure theory of interest rate is divided into traditional term structure theory, static theory and dynamic theory. Firstly, this paper analyzes the background and significance of the research on the term structure of national debt interest rate, as well as the research situation of the term structure of interest rate at home and abroad. In the second chapter, the traditional term structure theory of interest rate and the theory of term structure model of interest rate are systematically introduced. In the third chapter, the parameter estimation method of two-factor CIR model is described in detail, that is, Kalman filter. In the part of empirical research, the Nelson-Siegel model is used to generate the data of spot interest rate. By analyzing the correlation coefficient of interest rate in each period, it is concluded that the single factor model can not accurately describe the dynamic process of interest rate. Factor analysis shows that the two-factor model has been able to describe most of the dynamic characteristics of interest rates. Secondly, on the basis of spot interest rate, the parameters of two-factor CIR model are obtained by Kalman filter. Finally, it can be seen from the error root mean square value estimated by the two-factor CIR model that the fitting of the 1-year spot interest rate model is not accurate enough, but it can accurately depict other spot interest rate data with longer term. On the whole, the data obtained by the Kalman filtering method is close to the actual data. Therefore, the two-factor CIR model can accurately describe the dynamic process of the term structure of interest rate in China's treasury bond market.
【学位授予单位】：安徽财经大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;F224;F812.5

【参考文献】