论文部分内容阅读
以极值理论为基础的风险值度量方法是最近发展起来的最为有效的方法之一,但在传统的单纯采用极值理论的建模过程中对误差项假定为独立同分布的白噪声过程,会对应用极值理论估算风险价值产生一定误差。本文以上证指数和深证成指为例,利用ARIMA-GARCH模型捕获股票收益序列中的自相关和异方差现象,对该模型中残差的条件分布的合理假定进行了实证分析比较,然后利用极值理论对经过ARIMA-GARCH模型筛选过的残差进行极值分析,估算风险价值。
The extreme value theory based measure of risk value is one of the most effective methods developed recently. However, the error term is assumed to be an independent and identically distributed white noise process in the traditional modeling process using extreme value theory only. There is a certain error in applying the extreme value theory to estimate the VaR. This paper takes Shanghai Stock Exchange Index and Shenzhen Stock Exchange as an example, uses the ARIMA-GARCH model to capture the autocorrelation and heteroscedasticity in stock returns series, empirically analyzes and compares the reasonable assumptions of the conditional distribution of residuals in this model, The extreme value theory analyzes the extreme value of the residuals screened by the ARIMA-GARCH model to estimate the VaR.