A study of stock returns based on ARMA-GARCH model: Taking the CSI 300 Index as an example

This paper selects the daily return data of CSI 300 index from January 2, 2014 to September 30, 2020 study characteristics volatility. An ARMA(3,3) model was fitted log returns using R software, and ARCH effect found exist. The were then with an ARMA(3,3)-GARCH(1,1) model, fitting tested, finally short-term predicted. results empirical analysis show that series has such as non-normal distributivity, spikes thick tails, aggregation.