Prediction of Pork Price Based on PCA-BP Neural Network

Abstract

In the research of pork price forecasting, due to the strong nonlinear relationship between the fluctuation of pork price and complex influencing factors, the traditional forecasting model cannot measure the nonlinear relationship and make an accurate prediction of pork price. To solve these problems, we propose a PCA-BP Neural Network prediction model to predict the price of pork. Firstly, the main factors affecting the fluctuation of pork prices are analyzed. 162 groups of data are used, including the national average weekly price of pork, white striped chicken, beef, mutton, corn, and soybean from the first week of January 2018 to the first week of February 2021. Three principal components with a 96% contribution rate are used as the input layer data of the BP neural network, and pork price is selected as the output layer data of the BP neural network. By comparing the predicted value with the actual value, the predicted value of the PCA-BP Neural network model is close to the actual value, and it has a better fitting effect and accuracy than the traditional BP neural network. The results show that the PCA-BP Neural Network pork prediction model provides new ideas for pork price prediction, which is of great significance to stabilizing the daily life of urban and rural residents and protecting the income of farmers.