基于MCRR测度中国股市风险的方法研究
发布时间：2019-05-22 05:01
【摘要】：市场经济不断发展,人们生活中必不可少的就是投资理财,随着市场经济的日趋发展以及受国外投资策略的影响,人们投资理财意识也不断提高,人们总是希望通过投资理财获得高于银行利率的收益,但是投资有风险,其实金融市场是有规律可循的,只要能够合理地测度金融风险,就可以一定程度上规避风险,从而减少一些不必要的损失。因此进行有效的金融监管具有重要的现实意义。金融资产的风险管理是金融监管的重要组成部分,要想进行有效的金融监管必须做好金融风险管理工作。 金融时序数据多数表现出不稳定的特点,资产收益率存在一定的波动性,可以用波动性来衡量资产的风险。这种波动呈现一种趋势：幅度较大的波动会相对聚集在某一段时间里,而幅度较小的波动会聚集在另一段时间里。本文针对金融资产的这种特性,运用MCRR方法研究我国股市风险,样本选取的是2008年1月到2013年3月上证指数和深圳成分指数日收益率。 本文共分为五个部分。第一部分是本文的引言部分,主要说明本文的研究背景和研究意义、文献综述、研究思路及本文的创新与不足之处；第二部分介绍有关的模型理论知识,包括ARCH族模型(ARCH、GARCH、TGARCH、 EGARCH等)的具体含义和作用,还有测度风险的两种方法——VaR和MCRR方法,分别介绍了两者的定义、计算方法和检验方法；第三部分建立了各种GARCH模型,主要是为了测度风险做准备,包括数据的描述性统计分析、平稳性检验、自相关检验,GARCH族模型的估计等；第四部分给出了VaR和MCRR方法测度风险的结果及检验；第五部分给出本文的结论和建议。 本文实证结果显示,与VaR方法相比较,MCRR方法有很多的优点。 首先,计算MCRR的估计方法更加精确。估计VaR的三种方法有参数法、历史模拟法和蒙特卡罗模拟方法。其中参数法依赖于收益率分布的假定,历史模拟法严重依赖历史数据,虽然蒙特卡罗模拟方法被认为是最全面和灵活的方法,但是该方法的模拟过程依赖特定的随机过程而不能保证准确模拟资产价格变化的过程,可能会导致不能准确地测度金融风险。运用Bootstrap方法计算MCRR,能够保证模拟资产价格变化的随机性,尤其能够考虑到金融时序数据的厚尾特征,从而准确测度金融风险。 其次,MCRR方法不仅能够测度风险的大小还能给出风险的范围。 最后,MCRR方法的优势还体现在GARCH模型上。从计算VaR的各种GARCH模型发现,这些模型都没有考虑模型中存在的“过度的波动持续性”,当把这种特性考虑进来时,即用GARCH-ONV模型模拟金融资产波动性,发现与其他GARCH模型比较,GARCH-ONV模型的衰减系数较小。关于风险测度结果比较,也发现GARCH-ONV估计的风险较小。这些都说明了其他GARCH模型中存在的“过度的波动持续性”,导致风险的高估,从而需要更高的资本要求。所以运用GARCH-ONV模型模拟金融资产的波动性,进行风险测度更准确。 因此,本文建议使用MCRR方法测度金融资产风险。进行有效的金融风险管理能够为经济主体提供一个安全稳定的经营环境,从而促进经济发展；进行有效的金融风险管理能够减少经济危机的发生,有利于社会的稳定。
[Abstract]:With the development of the market economy, people's life is essential to the investment and financing. With the development of the market economy and the influence of the foreign investment strategy, people's investment and financial management consciousness is also rising, and people always want to obtain the income higher than the bank interest rate through the investment finance, But the investment has the risk, in fact, the financial market is regular, so long as the financial risk can be measured reasonably, the risk can be avoided to a certain extent, so that some unnecessary losses can be reduced. Therefore, effective financial supervision is of great practical significance. The risk management of financial assets is an important part of financial supervision. Most of the financial time sequence data shows the characteristics of the instability, the rate of return of the assets has a certain volatility, and the wind of the assets can be measured with the volatility. Insurance. This kind of fluctuation presents a tendency: a large wave of amplitude will be relatively concentrated in a certain period of time, while the smaller amplitude will gather for another period of time In the light of this characteristic of financial assets, this paper uses the MCRR method to study the stock market risk in China. The samples are selected from January 2008 to March 2013 and the Shenzhen component index daily income. Rate. This article is divided into five. The first part is the introduction part of this paper, which mainly explains the research background and research significance, the literature review, the research thinking and the innovation and the deficiency of this paper. The second part introduces the relevant model theory, including ARCH, GARCH, TGA. The specific meaning and function of RCH, EGARCH, etc., and two methods of measure risk, VaR and MCRR method, are introduced in this paper. The definition, calculation method and test method of the two are introduced. The third part has set up a variety of GARCH models, mainly for the preparation of the measure risk, including the descriptive system of data. The fourth part gives the results and tests of the measure risk of the VaR and the MCRR method, and the fifth part gives the conclusion of this paper. The empirical results of this paper show that the MCRR method can be compared with the VaR method. A lot of advantages. First, calculate the estimate of the MCRR The method is more accurate. Three methods of estimating the VaR have the parameter method, the history simulation method and the control method. The Monte Carlo simulation method, in which the parameter method relies on the assumption that the yield distribution is distributed, the historical simulation method is heavily dependent on the historical data, although the Monte Carlo simulation method is considered the most complete Surface and flexible methods, but the simulation process of the method relies on a particular random process to ensure that the process of accurately simulating the change in the asset price may lead to an inability to accurately The financial risk is measured. The MCRR is calculated by the Bootstrap method, which can guarantee the randomness of the change of the price of the simulated asset, especially the thick-tail feature of the financial time sequence data can be taken into account. Second, the MCRR method can not only measure the size of the risk The scope of the risk is also given. Finally, the advantage of the MCRR method also The GARCH model has not taken into account the "over-fluctuation persistence" present in the model. The GARCH-ONV model is used to model the volatility of the financial assets, and the GARCH-O model is found to be compared with other GARCH models. The attenuation coefficient of the NV model is small. As for the comparison of the risk measure results, GARCH is also found. -The risk of the ONV estimate is small. These all account for the "over-fluctuation persistence" present in other GARCH models, resulting in an overestimation of the risk, Therefore, the GARCH-ONV model is used to model the fluctuation of the financial assets. The risk measure is more accurate. Therefore, it is recommended to use M The CRR method measures the risk of financial assets. Effective financial risk management can provide a safe and stable operating environment for the economic subject, thus promoting economic development, and carrying out effective financial risk management to reduce economic risk
【学位授予单位】：东北财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224
本文编号：2482698
[Abstract]:With the development of the market economy, people's life is essential to the investment and financing. With the development of the market economy and the influence of the foreign investment strategy, people's investment and financial management consciousness is also rising, and people always want to obtain the income higher than the bank interest rate through the investment finance, But the investment has the risk, in fact, the financial market is regular, so long as the financial risk can be measured reasonably, the risk can be avoided to a certain extent, so that some unnecessary losses can be reduced. Therefore, effective financial supervision is of great practical significance. The risk management of financial assets is an important part of financial supervision. Most of the financial time sequence data shows the characteristics of the instability, the rate of return of the assets has a certain volatility, and the wind of the assets can be measured with the volatility. Insurance. This kind of fluctuation presents a tendency: a large wave of amplitude will be relatively concentrated in a certain period of time, while the smaller amplitude will gather for another period of time In the light of this characteristic of financial assets, this paper uses the MCRR method to study the stock market risk in China. The samples are selected from January 2008 to March 2013 and the Shenzhen component index daily income. Rate. This article is divided into five. The first part is the introduction part of this paper, which mainly explains the research background and research significance, the literature review, the research thinking and the innovation and the deficiency of this paper. The second part introduces the relevant model theory, including ARCH, GARCH, TGA. The specific meaning and function of RCH, EGARCH, etc., and two methods of measure risk, VaR and MCRR method, are introduced in this paper. The definition, calculation method and test method of the two are introduced. The third part has set up a variety of GARCH models, mainly for the preparation of the measure risk, including the descriptive system of data. The fourth part gives the results and tests of the measure risk of the VaR and the MCRR method, and the fifth part gives the conclusion of this paper. The empirical results of this paper show that the MCRR method can be compared with the VaR method. A lot of advantages. First, calculate the estimate of the MCRR The method is more accurate. Three methods of estimating the VaR have the parameter method, the history simulation method and the control method. The Monte Carlo simulation method, in which the parameter method relies on the assumption that the yield distribution is distributed, the historical simulation method is heavily dependent on the historical data, although the Monte Carlo simulation method is considered the most complete Surface and flexible methods, but the simulation process of the method relies on a particular random process to ensure that the process of accurately simulating the change in the asset price may lead to an inability to accurately The financial risk is measured. The MCRR is calculated by the Bootstrap method, which can guarantee the randomness of the change of the price of the simulated asset, especially the thick-tail feature of the financial time sequence data can be taken into account. Second, the MCRR method can not only measure the size of the risk The scope of the risk is also given. Finally, the advantage of the MCRR method also The GARCH model has not taken into account the "over-fluctuation persistence" present in the model. The GARCH-ONV model is used to model the volatility of the financial assets, and the GARCH-O model is found to be compared with other GARCH models. The attenuation coefficient of the NV model is small. As for the comparison of the risk measure results, GARCH is also found. -The risk of the ONV estimate is small. These all account for the "over-fluctuation persistence" present in other GARCH models, resulting in an overestimation of the risk, Therefore, the GARCH-ONV model is used to model the fluctuation of the financial assets. The risk measure is more accurate. Therefore, it is recommended to use M The CRR method measures the risk of financial assets. Effective financial risk management can provide a safe and stable operating environment for the economic subject, thus promoting economic development, and carrying out effective financial risk management to reduce economic risk
【学位授予单位】：东北财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224
【参考文献】
中国期刊全文数据库 前10条
1 辛海明;;AR-GARCH模型在股票市场的应用[J];赤峰学院学报(自然科学版);2011年02期
2 田新时,刘汉中,李耀;沪深股市一般误差分布(GED)下的VaR计算[J];管理工程学报;2003年01期
3 刘庆富;仲伟俊;华仁海;刘晓星;;EGARCH-GED模型在计量中国期货市场风险价值中的应用[J];管理工程学报;2007年01期
4 康建林,朱开永,周圣武,韩苗;GARCH模型在中国股票波动预测中的应用[J];赣南师范学院学报;2005年03期
5 魏宇;;沪深300股指期货的波动率预测模型研究[J];管理科学学报;2010年02期
6 刘璐;张倩;任召敏;;后危机时代亚洲地区股票市场风险研究——基于非对称GARCH模型[J];经济与管理;2011年01期
7 徐建军;;GARCH族模型在汇率波动分析中的应用[J];经济师;2011年05期
8 陈守东,俞世典;基于GARCH模型的VaR方法对中国股市的分析[J];吉林大学社会科学学报;2002年04期
9 金素;刘璐;;GARCH模型在金融风险测度中的应用——以2005～2010年沪深指数为例[J];金融教学与研究;2011年03期
10 江涛;;基于GARCH与半参数法VaR模型的证券市场风险的度量和分析:来自中国上海股票市场的经验证据[J];金融研究;2010年06期
，本文编号：2482698
本文链接：https://www.wllwen.com/jingjilunwen/zbyz/2482698.html
