基于ARMA-GARCH模型创业板市场风险VaR和ES实证研究

发布时间：2018-04-17 13:38

本文选题：VaR + Expected　；参考：《南昌大学》2014年硕士论文

【摘要】：本文主要是讨论风险度量新方法ES的基本原理；基于ARMA-GARCH模型展开对我国创业板市场日收盘价对数收益率的市场风险实证分析，计算其VaR值和ES值来度量创业板指数的市场风险；此外，也运用向量自回归模型探讨创业板和中小板的相互影响，深度分析创业板的风险因素。全文共分5章。第一章介绍本文研究主要内容的背景、意义和方法，简要叙述VaR和ES相关历史文献。第二章主要介绍ES基本原理。采用资产组合损失变量描述风险，并基于损失分布的-（上）分位数给出“期望巨额损失值”ES（Expected Shortfall）和“条件风险价值”CVaR（Conditional Value at Risk）的定义。在一般损失分布下，通过直接计算说明了任一资产组合损失变量的“期望巨额损失值”ES的定义与-（上）分位数的选取无关；而且也通过直接计算证明了ES与CVaR两者的等价关系；进而通过构造出ES的概率测度族表示证明了ES是一致性风险度量方法。第三章基于ARMA-GARCH模型，运用风险度量的VaR和ES方法对创业板指数（样本：2010.6.1-2014.2.1）的市场风险进行实证分析，，包括VaR、ES的计算问题、比较分析等。首先对创业板块日收盘指数之间的自然对数收益率进行基本统计分析，发现序列的分布存在厚尾的特征；运用AIC+SC准则确定各模型的滞后阶数。对各模型的参数进行估计，得到残差序列，进一步得到残差序列的VaR值，并根据模型的数学表达式得到最终的VaR值和ES值。最后利用失败率检验法，对模型进行检验。实证的结果表明，ES在VaR估计失败时，能比较准确地估计例外情形的实际损失，即ES可以用来弥补VaR模型的缺点，ES是比VaR更加稳健和保守的风险度量方法。第四章运用VAR（向量自回归）模型，研究创业板收盘指数和中小板收盘指数的相互影响。通过协整以及格兰杰因果检验，分析创业板和中小板之间的内在关系，寻找影响创业板风险的外部因素。结果表明，它们的波动性是比较一致的；中小板和创业板之间不存在长期均衡关系。研究结果进一步表明，创业板对中小板的影响要大一些。第五章是对全文进行的总结并指出了文中的不足以及对今后研究方向的展望。
[Abstract]:This paper mainly discusses the basic principle of the new risk measurement method es, analyzes the market risk based on ARMA-GARCH model, calculates the VaR value and es value to measure the market risk of gem.In addition, the vector autoregressive model is used to study the interaction between the gem and the small and medium-sized board, and the risk factors of the gem are analyzed in depth.The full text is divided into five chapters.The first chapter introduces the background, significance and methods of the research, and briefly describes the historical documents related to VaR and es.The second chapter mainly introduces the basic principle of es.The risk is described by portfolio loss variables, and the definitions of "ES(Expected short fallings" and "CVaR(Conditional Value at risk" are given based on the-(upper) quantiles of the loss distribution.Under the general loss distribution, the definition of "expected huge loss value" es of any portfolio loss variable is directly calculated, which is independent of the selection of-(upper) quantiles.Furthermore, the equivalent relation between es and CVaR is proved by direct calculation, and then es is proved to be a consistent risk measure by constructing the representation of the probability measure family of es.The third chapter is based on the ARMA-GARCH model, using the VaR and es methods of risk measurement to analyze the market risk of gem index (sample:: 2010.6.1-2014.2.1), including the calculation problem and comparative analysis.Firstly, the natural logarithmic return rate between the daily closing index of entrepreneurial plate is analyzed, and it is found that the distribution of the sequence has the characteristic of thick tail, and the lag order of each model is determined by using AIC SC criterion.The parameters of each model are estimated and the residual sequence is obtained. The VaR value of the residual sequence is further obtained and the final VaR value and es value are obtained according to the mathematical expression of the model.Finally, the failure rate test method is used to test the model.The empirical results show that when the VaR estimation fails, it can accurately estimate the actual loss of exceptional cases, that is, es can be used to compensate for the shortcomings of VaR model and that es is a more robust and conservative risk measurement method than VaR.The fourth chapter studies the interaction between the gem closing index and the small and medium-sized board closing index by using VAR (vector autoregressive) model.Through cointegration and Granger causality test, this paper analyzes the internal relationship between gem and small and medium-sized board, and finds out the external factors that affect the risk of gem.The results show that their volatility is consistent, and there is no long-term equilibrium relationship between the small and medium-sized board and the growth enterprise board.The results further show that the gem has a greater impact on the small and medium-sized boards.The fifth chapter is a summary of the full text and points out the shortcomings of the paper and prospects for future research.
【学位授予单位】：南昌大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F224;F832.51

【参考文献】