参数法和非参数法VaR模型在股市风险中的比较研究——基于沪深300指数的实证研究

A comparative study on the application of parametric and non-parametric VaR models in stock market risk——Empirical Study Based on Shanghai-Shenzhen 300 Index

VaR作为风险度量运用最广泛的工具,测度方法也越来越得到投资者重视。在各种对于VaR的测度方法的概念与原理之上,基于多种不同的假设进行实证分析。研究得出,滑动窗口为250的加权历史模拟法在1%的置信度下,以及基于正态分布的NGARCH模型在5%的置信区间下,可以很好的对VaR进行测度。最后通过GARCH模型与EVT模型及蒙特卡洛模型的结合,提出了关于计算未来VaR新的度量方法,给予风险管理者更加直观的投资意见。

For risk measurement,as the most practical tool and its measurement method,VaR has been paid more and more attention by investors,recently. Based on a variety of different assumptions,empirical analysis concluded that the weighted historical simulation method with sliding window of 250 can measure VaR well at 1% confidence level and the NGARCH model based on normal distribution at 5% confidence level. Finally,through the combination of GARCH model,EVT model and Monte Carlo model,this paper puts forward a new measurement method for future VaR and gives risk managers more intuitive investment advice.