Research and Comparison of ARIMA and Grey Prediction Models for Subway Traffic Forecasting

With the increasing development of metro transportation in cities, the accurate prediction of station traffic poses a challenge for data mining and analysis. Machine learning has been gradually applied to various traffic prediction due to its low computational complexity, easy interpretation and accurate prediction results. In this paper, based on ARIMA and the gray prediction model, the data of a city's various subway gates are selected to predict the passenger flow of the subway. Then, the MAE (Mean Absolute Error) is used to compare the differences between the two machine learning algorithms on passenger flow prediction. Finally, this paper discusses the training set and model fitting to provide a technical reference for further improving the prediction performance.