Empirical study on the risk forecast of stock index futures and the CVaR-GARCH- GED Model

CVaR-GARCH-GED model can be used to describe the characters of the stock index futures return, such as peaks, thick tails and volatility clustering. The paper apply the CVaR-GARCH- GED model to the empirical study on the data sample collecting from daily income rate of stock index futures of Shanghai and Shenzhen 300 main contracts (IF1005).The conclusion is that fluctuations of CVaR forecast earnings based on GARCH-GED model is in compliance with the trend of the original returns. The accurate CVaR test reveals that the accuracy of CVaR forecast earnings under 95% confidence level is dramatic, that is, the CVaR-GARCH-GED model can predict the risk accurately.