基于MC-GARCH-VaR下的金融市场风险研究

发布时间：2018-05-23 06:54

本文选题：蒙特卡罗方法 + GARCH族　；参考：《山东大学》2012年硕士论文

【摘要】：随着股指期货和融资融券业务的开展,我国证券市场口趋发展和完善,股票市场作为整个国民经济的重要一环,其地位和作用也口益突出。而此时,我国金融行业还是处于一个转轨的阶段,货币市场和资本市场的波动都异常明显。在此特殊时刻,我国金融业的风险管理能力亟需加强,为了提升我国在金融市场中的竞争力,增强抵抗风险的能力,就必须掌握度量风险、预测风险的能力,进而对金融市场进行监控和管理。本文将借助个人在国内金融机构较丰富的实习经历,针对我国金融市场的现状,将传统VaR计算方法的计算步骤做具有现实性意义的阐述。这也是本文一大创新地方所在。 VaR方法在度量金融市场风险中已得到广泛的应用,而传统的VaR方法往往假设市场收益率服从正态分布。而这个假设在国外金融市场上已经验证为不符合实际。本文就我国2005年6月9日至2011年12月29日的HS300指数进行对数收益率分析,发现这个时间序列存在明显的尖峰厚尾性、异方差性等特性,于是考虑到用GARCH模型来模拟我国金融市场数据,作者分别用GARCH、EGARCH、TARCH、EGARCH-M、TARCH-M等GARCH族来模拟此金融时间序列,结果证实,就我国金融市场来说用GARCH(1,1)模型就能相对很好的进行模拟建模。然后利用蒙特卡罗方法进行金融序列的预测模拟,从而求出相应的VaR。并对此进行Kupiec回测检验,与之前用历史模拟法和正态模型法所作出的结果相比较,发现本文所用的MC-GARCH-VaR方法在功效上有了明显的提高,并且具有很强的稳定性。
[Abstract]:With the development of stock index futures and margin trading, the stock market is developing and improving. As an important part of the whole national economy, the stock market has a prominent position and function. At this time, China's financial industry is still in a transitional stage, the fluctuation of money market and capital market are very obvious. At this special moment, the risk management ability of our financial industry needs to be strengthened urgently. In order to enhance the competitiveness of our country in the financial market and strengthen the ability to resist the risk, we must master the ability to measure and predict the risk. Then the financial market monitoring and management. In this paper, with the help of individual practice experience in domestic financial institutions, and in view of the present situation of financial market in China, the calculation steps of traditional VaR calculation method are expounded with realistic significance. This is also a major innovation in this paper. VaR method has been widely used in the measurement of financial market risk, while the traditional VaR method often assumes the market yield from normal distribution. This assumption has been proven to be unrealistic in foreign financial markets. In this paper, the logarithmic rate of return of HS300 index from June 9, 2005 to December 29, 2011 is analyzed. It is found that this time series has obvious characteristics such as peak, thick tail, heteroscedasticity, etc. Therefore, considering the use of GARCH model to simulate the financial market data in China, the author uses GARCH family such as EGARCHM, EGARCH-MU TARCH-M to simulate the financial time series. The results show that the GARCH1) model can be used to simulate the financial market in China. Then Monte Carlo method is used to predict and simulate the financial series, and the corresponding VaR is obtained. Compared with the results obtained by the historical simulation method and the normal model method, it is found that the efficiency of the MC-GARCH-VaR method used in this paper is obviously improved and the stability is very strong.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.5;F224

【参考文献】