Research on the Return Rate of Shenzhen Composite Index in my country

Stock price forecasting plays an important role in judging the trend of my country's stock market. The research object is the daily closing price of Shenzhen Composite Index from January 2, 2015 to December 30, 2020. First, the ARMA (2,4) model is established for the logarithmic return sequence after the first-order difference. Secondly, through analysis, it is concluded that the residual sequence has conditional heteroskedasticity. First, we can try to study the GARCH (1,1) model in depth. Next, a comprehensive ARMA (2,4)-GARCH (1,1) model is constructed, and relevant diagnostic tests are performed. The research results show that the logarithmic return sequence of Shenzhen Composite Index has volatility clustering, and the short-term prediction effect is better than the long-term prediction effect.