基于GARCH类模型的VaR计算在我国金融市场中的实证研究

发布时间：2018-01-09 00:30

本文关键词：基于GARCH类模型的VaR计算在我国金融市场中的实证研究　出处：《暨南大学》2014年硕士论文　论文类型：学位论文

【摘要】：在经济飞速发展的今天，越来越多的金融衍生品和金融工具出现，金融市场对风险研究的要求也越来越高。风险度量是风险管理中的关键环节，其度量方法也在不断地改进和创新。VaR（Value at Risk）方法是度量金融风险的有效方法之一，在国内外金融机构中广泛应用，这一方法在某种程度上弥补了许多风险度量方法的不足。本文从金融市场风险的理论框架出发，就风险度量法中的VaR方法进行详细介绍，概述了VaR方法的基本原理、特点和计算步骤，详述了三种主要计算方法，并对VaR方法进行了评价分析。而后介绍了GARCH模型和它的拓展模型，以及在模型中处理序列尖峰后尾问题的方法。在实证分析中，选择上证综指日对数收益率序列作为研究对象，根据收益率序列在正态分布、t-分布和GED分布三种假定下分别建立GARCH(1,1)、EGARCH(1,1)和PARCH(1,1)模型，并对模型进行检验比较，得到合理结果。选择GED分布假设下的GARCH(1,1)、EGARCH(1,1)和PARCH(1,1)模型，分别计算不同置信水平下的VaR值，，运用Kupiec-失败频率检验法对模型进行检验。最终得出结论，在GED分布假设下的GARCH(1,1)模型是度量上证综指收益率风险的最优模型。
[Abstract]:In the rapid development of economy , more and more financial derivatives and financial instruments have emerged , and financial market demand for risk research is becoming higher and higher . Risk measurement is a key link in risk management , and its measurement method is continuously improved and innovated . VaR ( Value at Risk ) method is one of the effective methods to measure financial risk . VaR ( Value at Risk ) method is widely used in financial institutions at home and abroad . Starting from the theoretical framework of financial market risk , this paper gives a detailed introduction to VaR method in risk measurement method , summarizes the basic principle , characteristics and calculation steps of VaR method , details three main calculation methods , and evaluates VaR method . In the case of empirical analysis , we select the logarithmic yield sequence of SSE as the research object . The models are established under the assumption of normal distribution , t - distribution and GED distribution under the assumption of normal distribution , t - distribution and GED distribution . The VaR values at different confidence levels are calculated , and the models are tested by using the Kupiec - failure frequency test method . Finally , it is concluded that the models under the assumption of GED distribution are the best models to measure the risk of yield risk .

【学位授予单位】：暨南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.5;F224

【参考文献】