M-Copula-GARCH-t模型在股指期货风险管理中的应用

本文考虑到金融资产收益率这一时间序列的方差时变特性和尖峰厚尾的分布特征,所以使用GARCH-t模型来描述这一现象;同时为了分析包括股指期货在内的投资组合各项资产之间的相关结构,采用了不限制边缘分布的Copula方法,并最终提出了M-Copula-GARCH-t模型用以度量包括股指期货在内的投资组合各项资产之间的相关性,验证了Copula方法,特别是M-Copula方法在度量结果和拟合效果上,具有较高的精确性.

Application of M-Copula-GARCH-t Model in Risk Management of Stock Index Futures

In the article,taking into account that the variance of financial assets yield time series possess time-varying and fat tail distribution characteristics,we apply GARCH-t model into describing this phenomenon;meanwhile,in order to analyze the correlation structure of various assets among the portfolio including index futures,we have adopted a Copula method with no restrictions to the marginal distribution and ultimately proposed the M-Copula-GARCH-t model,which is used to measure the correlation of various assets among the portfolio including index futures.In the end,by comparing the effects with Gaussian model and Copula model,the measurement results and fitting effects verify that Copula method,especially M-Copula method has a more superior accuracy.