基于POT模型的股票市场风险度量研究

Research on Stock Market Risk Measurement Based on POT Model

股票市场风险度量是规避股市风险的关键所在,文章基于广义帕累托分布的POT模型对股票市场风险度量进行了研究。针对POT模型中阈值确定困难的问题,在对峰度法、Hill估计法分析的基础上,提出了斜率变点检测的阈值确定方法。研究发现,将POT模型应用于股票市场分析中,95%CI与90%CI相比,VaR更具可信度,且股市的流动性较低,其市场风险也就越大。

Stock market risk measurement is the key to avoid stock market risk.This paper studies the stock market risk measurement based on the pot model of generalized Pareto distribution.In view of the difficulty in determining the threshold in pot model,a threshold determination method for slope change point detection is proposed based on the analysis of kurtosis method and hill estimation method.Applying the pot model to the stock market analysis,the results show that the VaR value under the 95%confidence level is more reliable than that under the 90%confidence level.The smaller the liquidity of the stock market,the greater of the market risk.