GARCH模型与VaR的度量研究  

Empirical Analysis on GARCH-type Models and VaR

作　　者：徐炜[1] 黄炎龙[1] 

机构地区：[1]南京工业大学经济管理学院

出　　处：《数量经济技术经济研究》2008年第1期120-132,共13页Journal of Quantitative & Technological Economics

摘　　要：GARCH模型在金融资产序列波动率的模拟和金融风险VaR的度量中都有着广泛的应用。本文比较研究了RiskMetrics及GARCH族的11种模型分别在正态分布和Skewed-t分布下度量VaR值的精确程度,同时对向前一步预测的VaR值进行了失败率检测法和动态分位数测试。结果表明,Skewed-t分布较好地拟合了金融资产的厚尾特性;在不同的置信水平下,FIGARCH(BBM)、FIEGARCH及IGARCH模型预测的VaR值更加精确,其高估或低估的风险程度较轻。The model of GARCH is widely used in modeling the volatility of financial assets and measuring Va R. This paper comparatively studies RiskMetrics and GARCH-type models of 11, based on the assumption of gaussian normal distribution and skewed student＇s t distribution respectively and their accuracy of calibrating VaR. The study checks the one-step-ahead forecasting VaR by employing failure rate test and dynamic quantile test. The results show that skewed student＇s t distribution is better fitted with the feature of lepkurtosis and the models of FIGARCH （BBM）, FIEGARCH and IGARCH are more exactly than others, which the degree of high or low estimation is receivable.

关 键 词：VAR GARCH模型 Skewed-t分布 动态分位数测试 

分 类 号：F830[经济管理—金融学]