基于Nelson-Siegel参数类模型的利率期限结构研究

发布时间：2018-06-11 20:56

本文选题：利率期限结构 + Nelson-Siegel模型　；参考：《浙江工商大学》2017年硕士论文

【摘要】：近年来,随着中国金融市场的蓬勃发展,金融产品的种类不断增加,投资者数量也在大幅度增长,国债的大量发行使得国债在金融市场中的地位逐渐凸显,利率市场化进程的推进,使得利率期限结构的重要性逐渐凸显,利率期限结构已成为金融市场利率风险管理,金融资产的定价,货币政策制定的关键。因此,研究国债利率期限结构,为金融市场提供有价值的参考依据以及进行风险管理已成为重要的研究课题。本文在这一大背景下,首先对利率期限结构的相关理论和估计方法进行分析研究,发现参数类模型因其拟合效果好,参数的经济含义明显,模型比较稳定等优点受到了国内外研究学者的青睐,但是对于哪种模型更适合研究中国国债利率期限结构并没有得出统一的结论。为了研究哪种模型更适合国债市场,本文首先选取了 2012年1月到2017年3月的国债收益率月度数据,把这些数据按照到期期限分为1年到30年,并对2012年到2017年的利率期限结构进行了主成分分析,结果显示影响利率期限结构变动的因素有水平因素,倾斜因素,曲度因素,这些因素对利率期限结构曲线变动的方差贡献率分别为82.2002%,16.9948%,0.6283%,水平因素代表了利率期限结构的位置,倾斜因素代表了利率期限结构的倾斜程度,曲度因素决定了利率期限结构的弯曲程度,这为下文将要研究的Nelson-Siegel模型和Svensson模型做了铺垫,影响利率期限结构的变动的因素与Nelson-Siegel模型和Svensson模型的参数相对应,尤其是Nelson-Siegel模型的三个参数分别对应于水平因子,倾斜因子和曲度因子,说明Nelson-Siegel模型和Svensson模型可以很好进行国债利率期限结构的估计研究。为了对比研究Nelson-Siegel模型和Svensson模型的拟合效果,本文从第四章开始分别从横截面数据,样本内,样本外,模型参数的稳定性等四个方面对两模型进行了对比分析。实证结果表明,Nelson-Siegel模型的拟合效果好,估计的参数少,参数经济含义明显以及模型相对比较稳定,因此更适合进行国债利率期限结构的估计研究。
[Abstract]:In recent years, with the vigorous development of China's financial market, the variety of financial products has been increasing, and the number of investors has also increased by a large margin. The issuance of a large number of treasury bonds has made the status of national debt in the financial market gradually prominent. With the development of interest rate marketization, the importance of the term structure of interest rate is becoming more and more important. The term structure of interest rate has become the key to the management of interest rate risk, the pricing of financial assets and the formulation of monetary policy. Therefore, it has become an important research topic to study the term structure of government bond interest rate, to provide valuable reference for the financial market and to carry out risk management. Under this background, this paper first analyzes the relevant theories and estimation methods of term structure of interest rate, and finds that the parameter class model has obvious economic meaning because of its good fitting effect. The stability of the model has been favored by scholars at home and abroad, but there is no uniform conclusion as to which model is more suitable for studying the term structure of interest rate of Chinese government bonds. In order to study which model is more suitable for the bond market, this paper first selects the monthly data of bond yield from January 2012 to March 2017, and divides these data into 1 year to 30 years according to the maturity period. The principal component analysis of term structure of interest rate from 2012 to 2017 shows that the factors influencing the change of term structure of interest rate are level factor, inclination factor and curvature factor. The variance contribution rate of these factors to the change of term structure curve of interest rate is 82.2002 / 16.9948 and 0.62833.The horizontal factor represents the position of the term structure of interest rate, and the tilting factor represents the inclination degree of the term structure of interest rate. The curvature factor determines the degree of curvature of the term structure of interest rate. This paves the way for the Nelson-Siegel model and the Svensson model, which will be studied below. The factors affecting the change of the term structure of interest rate correspond to the parameters of the Nelson-Siegel model and the Svensson model. In particular, the three parameters of the Nelson-Siegel model correspond to the horizontal factor, the tilt factor and the curvature factor, respectively. It is shown that the Nelson-Siegel model and the Svensson model can be used to estimate the term structure of the interest rate. In order to compare and study the fitting results of Nelson-Siegel model and Svensson model, this paper makes a comparative analysis of the two models from four aspects: cross section data, sample data, outside sample and stability of model parameters. The empirical results show that the Nelson-Siegel model has good fitting effect, few parameters, obvious economic meaning and relatively stable model, so it is more suitable to estimate the term structure of treasury bonds.
【学位授予单位】：浙江工商大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F812.5

【参考文献】