基于时间序列的小波神经网络蔬菜价格预测模型

摘要

Aiming at an accurate prediction of vegetable price, the cabbage monthly price data and the relevant factors data in Hefei Zhou Gudui farm product market from 2005 to 2014 were collected as samples. By analyzing principal component and using the Wavelet Neural Network intelligent analysis method, the price forecasting model was built and was compared with BP neural network model. The results show that the wavelet neural network predictive model has higher precision and better stability than the BP neural network. The establishment of the model would provide technical support for stability of the price and references for related policies.%为准确预测蔬菜价格变化规律，现以合肥市周谷堆农产品批发市场2005年~2014年白菜月度市场价格及相关影响因素数据为样本，通过主成分分析，利用小波神经网络智能分析方法，构建基于小波神经网络的价格预测模型，并与BP神经网络模型比较。结果表明，小波神经网络预测模型的预测精度比BP神经网络更高，且更加的稳定。该模型的构建对蔬菜价格的稳定、农业管理部门的决策支持具有重要的理论研究意义和实际价值。