Neural Network Model Comparison and Analysis of Prediction Methods Using ARIMA and LSTM Models

Using the relevant data of the main corn futures contract of China Dalian Commodity Exchange from 2018 to 2021, the ARIMA model and the LSTM long short-term memory neural network model were established respectively, the two models were used to predict the daily closing price of corn futures, and compared with the actual. The prediction results of the two models are evaluated using the root mean square error (RMSE), average absolute error (MAE), and average error percentage (MAPE), the prediction capabilities of the two models are compared. The study found that the predictive ability of the LSTM model is better than the ARIMA model.