基于GARCH模型的利率与股价指数收益率波动相关性研究

发布时间：2018-04-22 18:58

本文选题：多元GARCH + 利率与股价指数收益率波动　；参考：《东北大学》2012年硕士论文

【摘要】：随着我国资本市场改革的深化和发展,股票市场在整个宏观经济中的作用日益突显。但是20世纪90年代以来,新兴市场国家爆发了一系列以股票市场相继崩溃为主要特征的金融危机,股票市场风险防范问题也尤为重要,各国货币当局对金融市场的宏观调控作用一直备受人们关注,如何控制与防范金融风险,也是摆在政府面前的一道难题。因此,作为货币政策重要的传导工具--利率,无疑在货币市场和股票市场的互动过程中起着不可替代的作用,研究二者相关性,可以为相关政策的制定提供参考信息,政策制定者可以利用利率与股票价格波动之间的相关性等信息统一调控金融市场,对于防范金融风险以及金融市场改革等方面的政策制定,具有重要的参考意义。 GARCH模型由于明确的经济涵义及对市场波动的准确刻画,得到了广泛应用,并证实了其有效性。同时,随着学者们的不断改进,衍生出许多新的GARCH类模型,利用时间序列工具对时变条件方差序列进行刻画。最近,又出现了研究多变量、多市场的多元GARCH模型,不仅涵盖了单变量的波动特性,又可刻画不同变量波动间的相关关系。但目前国内学者在研究股价和利率之间线性关系方面的论文较多,对二者之间波动关系的研究论文还比较少见。现实中,往往由于股市和利率对一些宏观政策等影响因素反应的异同,以及各自市场参与者投资行为的差异性等因素,使得二者之间关系更多地表现为非线性关联。因此,这就为采用多元GARCH方法研究中国的利率和股票市场波动相关性提供了新思路。为了检验中国的利率和股价指数收益率波动相关性,本论文在分析GARCH理论基础上,选取从2008年1月1日到2012年1月1日的近四年来的上海证券交易所综合股价指数收盘价日数据,应用一元GARCH族和多元GARCH模型,借助Eviews6.0和Matlab2010a软件进行了实证检验。根据研究重点的不同,论文将研究分为短期研究和长期研究：短期研究中,采用事件研究法,应用一元GARCH-M模型对收益率进行预测并计算超额收益率,通过对超额收益的显著性检验,考察了2008年1月至2012年1月期间的利率调整事件对上证指数收益率的短期影响,得到历次存贷款基准利率调整前后,上证综合股价指数波动并不显著的结论；在长期研究中,首先运用协整方法检验了利率与股价指数收益率之间的长期均衡关系,然后运用JB统计量和Q统计量对利率和股价指数收益率数据进行基本统计分析,检验残差序列的ARCH效应和条件方差的不对称性,并在此基础上运用二元EGARCH模型对利率与股票市场波动的相关性进行了实证分析,得出利率与股价指数之间存在双向波动溢出效应的结论。最后,根据中国的利率和股票市场波动相关性的实证检验结果,提出了相关政策建议。论文分六章展开：第一章,引言；第二章,GARCH族、利率与股价指数、波动及相关性备用知识；第三章,中国利率与股价指数波动相关性GARCH模型的建立；第四章,利率与股价指数波动长、短期相关性的实证检验；第五章,实证检验结果分析和政策建议；第六章,总结与展望。
[Abstract]:With the deepening and development of China's capital market reform, the role of the stock market in the whole macro-economy is becoming more and more obvious. But since 1990s, the emerging market countries have erupted a series of financial crises which are mainly characterized by the collapse of the stock market. The risk prevention of stock market is also particularly important. The macro-control role of the financial market has always been paid attention to. How to control and prevent financial risks is a difficult problem in front of the government. Therefore, as an important transmission tool of monetary policy, interest rate will undoubtedly play an irreplaceable role in the process of interaction between the money market and the stock market. It is possible to study the relevance of the two parties. In order to provide reference information for the formulation of relevant policies, policy makers can regulate the financial market by using the information between interest rate and the fluctuation of stock prices. It is of great reference significance to the policy formulation of financial risk prevention and financial market reform.
GARCH model has been widely used because of its clear economic meaning and accurate portrayal of market volatility. It has proved its effectiveness. At the same time, with the continuous improvement of the scholars, many new GARCH models are derived, and time series tools are used to characterize the time-varying conditional variance sequences. The multivariate GARCH model of the field not only covers the fluctuation characteristics of the single variable, but also portrays the correlation between the fluctuation of different variables. However, there are many papers on the study of the linear relationship between stock price and interest rate at present, and the research papers on the fluctuation relationship between the two are relatively rare. Some factors such as macro policy and other impact factors, as well as the differences in the investment behavior of the market participants, make the relationship between the two more nonlinear. Therefore, this provides a new idea for the study of China's interest rate and the volatility of the stock market by using the multiple GARCH method.
In order to test the correlation between China's interest rate and the volatility of stock index returns, this paper, based on the analysis of GARCH theory, selects the daily data of the closing price of the Shanghai stock exchange's comprehensive stock index from January 1, 2008 to January 1, 2012, using the GARCH and multiple GARCH models with the help of Eviews6.0 and Matlab2010a. According to the different research focus, the paper divides the research into short-term and long-term research. In the short term, the event study method is used to predict the rate of return by using the one yuan GARCH-M model and to calculate the excess return rate. The period from January 2008 to January 2012 is examined through the significant test of excess returns. The short-term effect of interest rate adjustment on the rate of return of Shanghai stock index has been concluded. In the long-term study, the long-term equilibrium relationship between interest rate and the rate of return on stock index is tested by cointegration method in the long term study, and then the JB statistics and the Q system are used. The basic statistical analysis of interest rate and stock index yield data is carried out to test the asymmetry of ARCH effect and conditional variance in the residual sequence. On this basis, the correlation between interest rate and stock market volatility is empirically analyzed by using two yuan EGARCH model, and it is concluded that there is a two-way volatility spillover effect between the profit rate and the stock index. Finally, according to the empirical test results of China's interest rate and stock market volatility, relevant policy recommendations are put forward.
The thesis is divided into six chapters: Chapter one, introduction; the second chapter, GARCH family, interest rate and stock index, volatility and correlation spare knowledge; the third chapter, the establishment of GARCH model of China interest rate and stock index volatility; the fourth chapter, the empirical test of the volatility of interest rate and stock index, and the short-term correlation; the fifth chapter, the empirical test result is divided into two chapters. Analysis and policy recommendations; the sixth chapter, summary and prospect.

【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;F224

【参考文献】