经济政策不确定性与中国股市波动率预测——在GARCH-MIDAS模型下的实证研究

Economic policy uncertainty and forecasts of Chinese stock market volatility——Empirical Research under the GARCH-MIDAS Model

文章以中国经济政策不确定性指数(CEPU)、美国经济政策不确定性指数(AEPU)和全球经济政策不确定性指数(GEPU),上证综合指数5分钟高频数据和日收益率为研究对象,运用GARCH-MIDAS模型分析不同经济政策不确定性对上证综合指数波动率的影响。实验结果表明,研究结果发现:GARCH-MIDAS模型比GARCH模型可以更好地拟合我国股市的波动状况,同时加入EPU指数能提升模型的预测性能。此外,通过不同的预测窗口和DM检验结果发现,CEPU对中国股票市场波动影响最大,说明本国经济政策是中国股票市场长期波动的主因。外国经济政策虽然会对本国股票市场波动产生影响,但其影响的程度远没有本国的经济政策对股市影响的效果强烈。

The article uses the China Economic Policy Uncertainty Index(CEPU), the US Economic Policy Uncertainty Index(AEPU) and the Global Economic Policy Uncertainty Index(GEPU), the 5-minute high-frequency data and daily yield of the SSE Composite Index were used as the research objects, and the GARCH-MIDAS model was used to analyze the impact of different economic policy uncertainties on the volatility of the SSE Composite Index. The experimental results show that the GARCH-MIDAS model can better fit the fluctuation of China's stock market than the GARCH model, and the addition of EPU index can improve the predictive performance of the model. In addition, through different forecast windows and DM test results, it is found that CEPU has the greatest impact on the volatility of China's stock market, indicating that the long-term fluctuations of China's stock market are mainly affected by policy changes. Although foreign economic policies will have an impact on the fluctuations of the domestic stock market, the impact of foreign economic policies on China's stock market is not as strong as the impact of domestic economic policies on the stock market.