论文部分内容阅读
本文从理论和实证两方面对目前流行的各类股票市场风险计量方法的特点和优劣进行比较分析。创新之处在于采用复合模型,用GARCH类模型改善VaR风险计量模型,并用Bootstrap方法修正模型。由于条件均值的ARMA方程允许模拟任何序列相关,而GARCH条件方差方程可应用于处理波动聚类,因此,采用GARCH类模型可得到独立同分布的观察值,得到更有效的VaR估计值。
This article compares and analyzes the characteristics and advantages and disadvantages of the currently popular stock market risk measurement methods both theoretically and empirically. The innovation lies in adopting the compound model, improving the VaR risk measurement model with the GARCH class model, and correcting the model with the Bootstrap method. Since the conditional mean ARMA equation allows any sequence correlation to be simulated, and the GARCH conditional variance equation can be used to deal with fluctuating clustering, using the GARCH class model results in independent and identically distributed observations, resulting in more efficient VaR estimates.