A hybrid improved LSTM-CNN model for Chinese stock price trend prediction

Predicting the trend of stock price is a challenging task and good evaluation models can bring huge profits. The classical linear prediction models are not suitable for the stock price trend prediction because the stock market is complex and dynamic. In order to accurately predict the trend of stock price, this paper proposes a hybrid and improved LSTM-CNN stock forecasting model and applied it to predicting the Chinese stock price trend. In addition, we further analyzed the correlation between Chinese stocks and network structure of Chinese stock market. The experimental results show that stocks which belongs to the same sectors influence each other in Chinese stock market, and the LSTM-CNN model has stable accuracy and low risk in the data sets, which shows it has stronger ability to predict stock price trend compared with RF, CNN, and LSTM model. These conclusions can guide investors to make reasonable decisions.