Modeling Opening Price Spread of Shanghai Composite Index Based on ARIMA-GRU/LSTM Hybrid Model

This study explores and models the 'opening price difference rate' of the Shanghai Stock Exchange Composite Index through exploratory data analysis and forecasting. The findings indicate that the hybrid models, ARIMA-LSTM and ARIMA-GRU, are better able to capture the trend and volatility of the data compared to single time series models or deep learning models. By combining the ARIMA model to capture the trend and LSTM/GRU models to capture the volatility, more accurate predictions can be obtained. These research results have significant practical implications for stock market forecasting and investment decision-making