基于半参数模型的风险度量方法及其应用

Risk Measurement Method Based on Semi-parametric Model and Its Application

目前度量预期不足的风险技术大多基于参数模型,其建模过程避免不了对收益的分布类型做出假定,但这些分布往往与现实相悖。鉴于此,文章提出两种重要半参数模型:CARE模型和CARES模型。应用我国2007—2016年上证综合指数与深证成分指数的相关数据评估模型优劣。结果表明:CARES模型与CARE模型在度量我国股市风险中都具有较好的效果,但两者比较,CARES模型明显优于CARE模型。因此,CARES模型能作为我国股市风险度量工具中的一个重要补充。

Most current methods for measuring expected shortfalls are based on parametric models,and the modeling process can’t avoid making assumptions about the distribution types of returns,but these distributions tend to be contrary to reality.In view of this,the paper proposes two important semi-parametric models:CARE model and CARES model,and applies the relevant data of China’s Shanghai Composite Index and Shenzhen Component Index from 2007 to 2016 to evaluate the advantages and disadvantages of the models.The results show that both the CARES model and the CARE model have a good effect in measuring the stock market risk in China,but compared with the CARE model,CARES model is obviously superior.Therefore,the CARES model can be used as the important risk measurement tools of China’s stock market.