我国主要粮食价格预测预警研究——基于神经网络及控制图理论分析

Research on Prediction and Early Warning of Major Grain Prices in China——Theoretical Analysis Based on Neural Network and Control Chart

本文结合前人研究和国外经验,以大豆为例,深入研究粮食价格预测预警机制。主要利用机器学习的Lasso方法及神经网络理论,结合灰色预测进行粮食价格预测的优化,结果表明预测精度高,模型可以很好地拟合大豆价格变动情况。在此基础上,引用控制图理论,结合预测价格,构建了一个有效的价格预测预警机制。本文建立对大豆价格预测预警机制具有可行性,一定程度上可以推广到其他主要粮食价格的预测预警研究中,可为保障我国粮食安全,保护农民的利益,保证粮食市场的稳定发展,为政府制定价格干预政策提供一定的借鉴。

In this paper, with the previous research and foreign experience, soybean as an example, in-depth study of grain price prediction and early warning mechanism. Lasso method and neural network theory are used to optimize grain price prediction combined with grey prediction. The results show that the prediction accuracy is high, and the model can well fit the price change of soybean. On the basis of this, the author constructs an effective price forecasting and warning mechanism based on the control chart theory and the forecast price. The forecasting and early warning mechanism of soybean price is feasible and can be extended to the forecast and early warn- ing research of other major grain prices. In order to provide some reference to help safeguard our country＇s grain security, protect the farmer＇s benefit, guarantee the stable development of the grain market, and draw up the price intervention policy for the government.