沪深股市的波动性分析——基于t分布下GARCH和SV模型的比较

Volatility Analysis of the Shanghai and Shenzhen Stock Market——The Compare of GARCH and SV Models Based on the t Distribution

构建基于N分布和t分布下的GARCH(1,1)和SV模型,并通过实证分析探讨了上证指数和深证成指收益序列的波动性.分析结果表明,GARCH(1,1)类模型和SV类模型能较好地拟合沪深股市收益率的波动,并指出我国股市存在较强的波动持续性;而基于t分布的各模型能有效地刻画股市的厚尾性;此外,通过计算VaR值,说明深市比沪市的风险更大,且SV类模型能更准确地反映收益率的风险特性.

The GARCH（1,1） of return volatilities of Shangh and SV models based on N distribution and t distribution were built, and the rate ai stock index and the Shenzhen component index were investigated through the empirical analysis. The results show that GARCH （1,1） and the SV model fit the volatility of Shanghai and Shenzhen stock market and the volatility of stock market is an ongoing process. Moreover,Shenzhen stock market is more risky than Shanghai based on VaR, and SV model can reflect the rate of return volatility more accurately.