基于参数拟合模型的交易所与银行间市场国债利率风险比较研究

发布时间：2018-02-14 17:53

本文关键词： 交易所银行间市场参数拟合模型利率风险 PCA　出处：《厦门大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着我国利率市场化进程的逐步推进,市场资金的供求关系逐渐成为影响利率的主要因素,而利率的波动也更加的频繁。如何对利率期限结构进行建模,并从中提取出利率波动的风险因素变得越来越重要。由于交易机制、参与主体上的差异,交易所与银行间债券市场形成两个分割的主体。本文基于利率期限结构理论,以交易所与银行间国债市场作为研究对象,运用参数拟合模型分别拟合两个市场的利率期限结构。根据估计得到的即期利率时间序列数据,运用主成分分析(PCA)方法提取利率波动的风险因素,解释各个因子所代表的涵义,并与美国国债市场进行对比,进而提出构建统一、高效的债券市场的对策。按照上述研究思路,本文所做的工作主要包括以下几个方面：第一,选取2012年1月6日至2013年12月3日上海证券交易所与银行间债券市场的国债原始数据进行实证分析,基于Nelson-Siegel(NS)模型和Svensson(SV)模型分别拟合了交易所与银行间国债市场即期利率的利率期限结构,通过理论分析和实证对比,发现SV模型更适合用来拟合我国国债利率期限结构。第二,基于SV模型得到的即期利率,通过主成分分析方法提取交易所国债市场与银行间国债市场各自的利率波动风险因素,对比分析两个市场利率波动的风险性因素,并与美国国债市场进行对比。通过对比研究,发现三因子利率动态模型基本上能够刻画美国国债市场的收益率变动,即水平因素(level)、倾斜因素(slope)和曲度因素(curvature);而中国市场(主要是交易所与银行间市场)需要四个因子才能较好地涵盖国债收益率变动的风险。其中,交易所与银行间市场第四个因子的方差贡献率分别达到了2.9849%和2.3868%,说明利率风险中未被解释的部分比重仍比较大。从市场分割的角度出发,作者认为交易机制、信息反应速度、债券发行规模、风险对冲机制是造成第四个因子比重较大的原因。第三,本文实证发现,相较于唐革榕和朱峰(2003)[1]的研究结果,交易所与银行间市场的国债利率期限结构对三因子模型的解释能力都提高了很多,而且两个市场之间的差距也在不断缩小。相比之下,银行间债券市场的利率波动分解出的倾斜因素和曲度因素解释总方差变动的比率与美国债券市场的情况比较接近,这可能与银行间市场的交易机制、参与主体跟国外市场比较相似有关。
[Abstract]:With the development of interest rate marketization in China, the relationship between supply and demand of market funds has gradually become the main factor affecting interest rate, and the fluctuation of interest rate is more frequent. How to model the term structure of interest rate, It is becoming more and more important to extract the risk factors of interest rate fluctuation. Because of the trading mechanism and the difference of the participants, the exchange and the interbank bond market form two separate subjects. This paper based on the term structure of interest rate theory, Taking the market of exchange and interbank bonds as the research object, the paper uses the parameter fitting model to fit the term structure of interest rate of the two markets, and according to the estimated time series data of spot interest rate, The principal component analysis (PCA) method is used to extract the risk factors of interest rate fluctuation, to explain the meaning of each factor, and to compare it with the US Treasury bond market, and then put forward the countermeasures to construct a unified and efficient bond market. According to the above research ideas, the work done in this paper mainly includes the following aspects:. First, the original data of the Shanghai Stock Exchange and the interbank bond market from January 6th 2012 to December 3rd 2013 are selected for empirical analysis. Based on the Nelson-Siegeler NSmodel and the Svenssonn model, the term structure of the spot interest rate in the exchange market and the interbank bond market is fitted, respectively. Through theoretical analysis and empirical comparison, it is found that SV model is more suitable for fitting the term structure of the interest rate of China's treasury bonds. Secondly, based on the spot interest rate obtained by SV model, the risk factors of interest rate fluctuation in the exchange bond market and the interbank bond market are extracted by principal component analysis, and the risk factors of interest rate volatility in the two markets are compared and analyzed. Through the comparative study, it is found that the three-factor interest rate dynamic model can basically depict the change of the yield of the US Treasury bond market. That is, horizontal factors, slope factors and curvature factors, while the Chinese market (mainly exchanges and interbank markets) needs four factors to better cover the risk of bond yield movements. The variance contribution rate of the fourth factor in the exchange and interbank market has reached 2.9849% and 2.3868 respectively, indicating that the unexplained proportion of the interest rate risk is still relatively large. The scale of bond issuance and the risk hedging mechanism are the reasons for the higher proportion of the fourth factor. Thirdly, this paper finds that compared with the results of Tang Guanyong and Zhu Feng 2003 [1], the term structure of the bond interest rate in the exchange and interbank markets has greatly improved its ability to explain the three-factor model. And the gap between the two markets is narrowing. In contrast, the interest rate volatility in the interbank bond market decomposes the slope factor and the curvature factor to explain the change of total variance, which is similar to that in the US bond market. This may be related to the interbank market trading mechanism, the participants and foreign markets are more similar.
【学位授予单位】：厦门大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51;F224

【参考文献】