基于PCA-GM-BP神经网络的猪肉价格预测分析

Analysis of Pork Price Prediction Based on PCA-GM-BP Neural Network

针对猪肉价格上下波动呈非线性关系和影响因素复杂等难以预测的问题,提出了基于PCA-GM-BP神经网络预测模型对猪肉价格进行有效预测.以2010年1月-2018年12月的月度价格数据作为样本,共计108组数据,利用PCA对影响猪肉价格变化的12种因素进行降维处理,选用对猪肉价格的主要累积贡献率超过96%的5个主成分,构建PCA-GM-BP神经网络猪肉价格预测模型.结果表明:与传统的BP神经网络、GM-BP神经网络预测模型相比,PCA-GM-BP神经网络预测模型在提高聚类效果的同时,增加了预测结果的精确性,对我国猪肉价格预测具有更高的适用性与参考价值性.

Aiming at the unpredictable problems such as the non-linear relationship between the up and down fluctuations of pork prices and the complex influencing factors,this article proposes an effective prediction of pork prices based on the PCA-GM-BP neural network prediction model.Taking monthly price data from January 2010 to December 2018 as a sample,a total of 108 sets of data,using PCA to reduce the dimensionality of 12 factors that affect pork price changes,and selecting the main cumulative contribution rate of pork prices to exceed 96% Based on the five principal components,the PCA-GM-BP neural network pork price prediction model was constructed.The results show that:Compared with the traditional BP neural network and GM-BP neural network prediction model,the PCA-GM-BP neural network prediction model improves the clustering effect while increasing the accuracy of the prediction results,which can predict pork prices in China.It has higher applicability and reference value.