Egg Price Forecasting in Shandong Province Based on PCA-GWO-LSTM Model

Shandong Province is a major producer of eggs. Forecasting egg prices in advance is of great significance for promoting the healthy development of the egg and poultry industry in Shandong Province. Therefore, a prediction model of egg prices based on principal component analysis combined with grey wolf optimization algorithm for long short-term memory neural networks is proposed. In order to reduce the computational complexity of the model while improving its prediction accuracy, the principal component analysis method is first used to identify the most significant influencing factors among the many factors affecting egg prices, achieving the purpose of data dimensionality reduction. Then, the grey wolf optimization algorithm is utilized to adjust and optimize the parameters of the long short-term memory network model, searching for the optimal model parameters and constructing the PCA-GWO-LSTM model. Finally, the model is validated using experimental data from weekly egg price data from January 2014 to October 2023 for training and testing. The results indicate that the model can effectively reflect the trend of egg prices in Shandong Province and has a good fitting effect. In terms of model error values, the average absolute error, coefficient of determination, and root mean square error are 0.236, 1.12, and 0.284. Compared with the GRU neural network model, in terms of the coefficient of determination indicator, this model improves accuracy by 6.67% over the GRU model. The model established in this study demonstrates high accuracy and can accurately forecast egg prices in Shandong Province. It can provide decision support for egg price monitoring in Shandong Province, thus offering significant practical application value.