摘要
VaR能够方便地度量金融市场的风险水平,已成为众多银行和金融机构监管与防范市场风险的标准工具。VaR模型是否稳定是实际使用者所关注的十分重要的问题。本文介绍了诺贝尔经济学奖得主Engle及合作者Manganelli于1999年最新提出的条件自回归VaR模型(简记CAViaR模型)的理论及实际意义,并采用Chow检验方法重点探讨了中国股市风险CAViaR建模的稳定性问题。通过对上证综合指数与B股指数及深圳成指的实证研究,我们发现,由于中国股市的结构性变化,Engle及Man-ganelli所力荐的四个参数CAViaR模型用于中国股市风险建模是相当不稳定的,不能很好地适合现实的中国股票市场。忽略中国股市的结构性变化而度量市场风险会低估或高估模型的解释能力。
Value-at-Risk (VaR) has become a standard tool to measure market risk, widely employed by financial institutions both for internal and regulatory purposes. The stability of a VaR modeling is usually important for prediction and econometric inference in practice. In this paper, the CAViaR model, proposed recently by Engle and Manganelli (1999) to calculate VaR, is introduced, and the Chow′s test for parameter instability is applied to explore the model stability of CAViaR modeling of the risk of Chinese stock markets. The empirical results show that the four popular CAViaR models highly recommended by Engle and Manganelli do not fit well the real situation of Chinese stock markets. Negligence of Chinese stock markets′ structural changes may lead to reduction of the interpretability in modeling of market risk.
出处
《管理评论》
2004年第5期9-16,共8页
Management Review
基金
国家自然科学基金资助。
