A Deep Model of Futures Contracts Based on Multi contract Collaborative Prediction

At present, the prediction effect of using deep learning to predict futures prices is usually not good. A Multi Contract LSTM model (MC-LSTM) was designed and implemented to predict futures prices by utilizing actual futures trading data and utilizing the Recurrent Neural Network (RNN) as the basic architecture, as well as multi contract fusion technology. By merging data features and optimizing parameters, the prediction accuracy of the algorithm has been improved compared to other similar algorithms. The work provides a decision-making reference for participants in the futures market.