Pork Price Prediction Using Bi-RNN-LSTM Artificial Neural Network

Abstract

Pork is one of the main methods for human intake of animal protein, and its price level will directly affect people’s daily lives. In order to realize the prediction of the prices in the live pig (mid-term) market, based on monthly data provided by China National Database, in this paper we propose a combination of artificial neural network models based on bidirectional recurrent neural network and bidirectional long short-term memory as the backbone network. The prediction errors achieved on our data set for Mean Square Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Symmetric Mean Absolute Percentage Error (SMAPE) are 0.48, 0.69, 0.53, 3.37%, 3.37% respectively. Compared with other deep learning models, the error of this method is small, which shows that it has the ability to predict the time series of pork price.