基于EGARCH-GED模型下的股市风险测度研究

Study on Risk Measurement in Chinese Stock Market Based on EGARCH-GED Model

根据我国股市市场收益的基本特征,首先运用AR-EGARCH模型来捕获上海证综合指数收益序列的自相关性、波动集聚性和杠杆效应;然后利用广义误差分布估计其厚尾分布,建立了能准确度量时变风险价值的AR-EGARCH-GED模型,并与基于正态分布和学生t分布的AR-EGARCH模型所计算的风险价值效果进行比较.最后,通过实证分析,并利用后验测试,表明基于AR-EGARCH-GED模型的风险价值能更好地刻画我国股市的市场风险.

Firstly, based on the basic characteristics of return series in Chinese Stock Market, AR - EGARCH model is used to obtain its effect of autocorrelation, clustering fluctuation and especially the ＂leverage effect＂ of fluctuation. Generalized error distribution is then adopted to estimate the fat tail of return sequences. The AR- EGARCH - GED model for calculating value at risk is established and then compared with the AR - EGARCH models based on normal distribution and t distribution. Finally, an empirical analysis in combination with posterior testing shows that the AR-EGARCH -GED model is a better method to describe the market risk in Chinese Stock Market.