引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
过刊浏览    高级检索
本文已被:浏览 36次   下载 0  
分享到: 微信 更多
小波变换-BP神经网络的农产品价格预测研究
刘合兵, 韩晶晶, 席磊
河南农业大学信息与管理科学学院
摘要:
【目的】农产品价格变动关乎国计民生,由于农产品的价格受到多方面因素的共同影响,其价格预测也一直是研究中的难点。通过对农产品价格进行组合预测模型的短期精准预测,指引农产品产业健康发展。【方法】文章根据农产品价格的特性以菠菜、大白菜、西红柿、辣椒和土豆五种蔬菜为代表研究其价格变动趋势。通过选用2013年1月到2018年12月的共72组月度价格数据,结合小波变换和BP神经网络两种方法构建农产品价格组合预测模型。首先利用小波变换对价格进行db5的3尺度分解,进而用BP神经网络模型对分解出的趋势部分和细节部分分别进行预测,最终对各分量的预测结果进行组合重构。【结果】预测精度的评价指标对五种蔬菜的价格预测结果进行对比分析,其平均绝对误差最小值为0.083元/kg,平均百分比误差最小为3.95%,均方根误差最小值为0.102。【结论】实验表明将小波变换和BP神经网络结合起来的组合预测模型对农产品的价格拟合较好,具有较高的预测性能,通过对多种蔬菜的价格验证证实该组合方法对农产品的价格预测具有普适性,且价格波动幅度和强度均会对其预测精度产生影响。
关键词:  小波变换;BP神经网络;农产品;价格预测
DOI:
分类号:
基金项目:1.河南省重大科技专项 2.河南省现代农业产业技术体系
Research on Agricultural Product Price Forecast Based on Wavelet Transform and BP Neural Network
Liu Hebing, Han Jingjing, Xi Lei
School of Information and Management Sciences, Henan Agricultural University
Abstract:
[Purpose] The price changes of agricultural products are related to the national economy and the people''s livelihood. Since the price of agricultural products is affected by many factors, its price forecast has always been a difficult point in research. Guide the short-term and accurate prediction of the combined forecasting model of agricultural products to guide the healthy development of the agricultural product industry. [Method] According to the characteristics of agricultural products, the article uses spinach, Chinese cabbage, tomato, pepper and potato as the representative of the price trend. By selecting a total of 72 sets of monthly price data from January 2013 to December 2018, combined with wavelet transform and BP neural network to construct agricultural product price combination forecasting model. Firstly, the 3 scale decomposition of db5 is carried out by using wavelet transform, and then the BP neural network model is used to predict the decomposed trend part and the detail part respectively. Finally, the prediction results of each component are combined and reconstructed. [Result] The forecasting accuracy evaluation index compares the price prediction results of five vegetables. The average absolute error is 0.083 yuan/kg, the average percentage error is 3.95%, and the root mean square error is 0.102. [Conclusion] Experiments show that the combined forecasting model combining wavelet transform and BP neural network has a good fit to the price of agricultural products and has high predictive performance. It is proved by the price verification of various vegetables that the combined method has a price forecast for agricultural products. Appropriateness, and price fluctuations and intensity will have an impact on its prediction accuracy.
Key words:  wavelet transform; BP neural network; agricultural products; price forecast