基于GARCH模型和SV模型的VaR比较  

Comparison of VaR based on GARCH and SV models

作　　者：余素红[1] 张世英[1] 宋军[1] 

机构地区：[1]天津大学管理学院,天津300072

出　　处：《管理科学学报》2004年第5期61-66,共6页Journal of Management Sciences in China

基　　金：国家自然科学基金资助项目(70171001).

摘　　要：简单介绍了VaR的含义及计算方法,指出推测市场因子的波动情况是计算VaR的关键.通过对比GARCH和SV模型,得出SV模型更能刻画金融市场的实际特征.将随机波动SV模型应用于VaR的计算,最后作实证研究.通过与GARCH模型下的结果对比,说明基于SV模型计算的VaR更具有动态性和准确性,VaR更贴切地反映了金融市场的风险水平.This paper reviews the concept and the calculating method of Value at Risk(VaR) and points out the importance of predicting the volatility of market factor for calculating VaR. Compared with GARCH models, SV model is superior to describe the characters of financial market. A SV model is used to define the volatility which is needed to estimate VaR. The experimental research manifest that the SV model predicts the volatility of market return perfectly and the following VaR reflect the risk level of Chinese stock market properly.

关 键 词：VAR 市场因子 GARCH模型 SV模型 

分 类 号：F830.91[经济管理—金融学]