江西省生猪价格波动的成因及其预警分析——基于灰色关联和LS-SVM模型

发布时间：2018-02-25 03:26

本文关键词： 生猪价格波动特征灰色关联 LS-SVM模型预警　出处：《浙江农业学报》2016年09期 　论文类型：期刊论文

【摘要】：以2000年1月至2015年5月的江西生猪价格数据为研究对象,利用Census-X12和HP滤波分解方法探索生猪价格波动的特征,结合逐步回归法和灰色关联分析识别影响生猪价格波动的显著因素,在此基础上,构建LS-SVM模型对生猪价格进行预测。结果表明,生猪价格波动具有明显的季节性,每年的1月份季节因子最大,6月份降至全年的最低点;2000年以来生猪价格共经历了7个波动周期,平均周期为25.3个月;随机性成分对生猪价格的贡献日益增大,玉米价格、仔猪价格、猪肉价格、生产者预期、牛肉价格和疫情对生猪价格的波动有显著作用,其中玉米价格和仔猪价格的影响较大;LS-SVM模型的预测值和真实值很接近,平均误差仅为1.37%,LS-SVM能较好地反映生猪价格及其影响因素之间的复杂的非线性关系。
[Abstract]:Taking Jiangxi live pig price data from January 2000 to May 2015 as the research object, using Census-X12 and HP filter decomposition method to explore the characteristics of live pig price fluctuation. Combined with stepwise regression and grey correlation analysis, the significant factors affecting the price fluctuation of live pigs were identified. On the basis of this, LS-SVM model was constructed to predict the price of live pigs. The results showed that the fluctuation of prices of live pigs had obvious seasonality. In January, the seasonal factor reached the lowest point in the whole year. The hog price has gone through seven cycles since 2000, with an average cycle of 25.3 months. The contribution of random ingredients to pig price is increasing, and the price of corn is increasing. Piglet price, pork price, producer expectation, beef price and epidemic situation have significant effects on the fluctuation of live pig price. The predicted value and real value of LS-SVM model are close to each other. The average error is only 1.37 and LS-SVM can well reflect the complicated nonlinear relationship between pig price and its influencing factors.
【作者单位】：江西农业大学理学院;江西农业大学江西现代农业发展协同创新中心;
【基金】：国家自然科学基金项目(71561014,61561025,71461019) 教育部人文社会科学研究项目(11YJCZH236) 江西现代农业及其优势产业可持续发展的决策支持协同创新中心课题(XDNYA1502,XDNYA1510)
【分类号】：F323.7;F224

【相似文献】