国内猪肉市场价格的EMD-SVM集成预测模型
详细信息 查看官网全文
- 英文论文题名：Integration Prediction of Domestic Pork Market Price Based on Empirical Mode Decomposition and Support Vector Machine
- 论文作者：蔡超敏 ; 凌立文 ; 牛超 ; 张大斌
- 英文论文作者：CAI Chao-min ; LING Li-wen ; NIU Chao ; ZHANG Da-bin ; College of Mathematics and Informatics ; South China Agriculture University ; Sun Point Century Technology Co.Ltd.
- 年：2016
- 作者机构：华南农业大学数学与信息学院;圣点世纪科技股份有限公司;
- 论文关键词：猪肉价格预测 ; 经验模态分解 ; 支持向量机 ; 集成预测
- 英文论文关键词：prediction of pork price ; empirical mode decomposition ; support vector machine ; integration prediction
- 会议召开时间：2016-11-12
- 会议录名称：第十八届中国管理科学学术年会论文集
- 英文会议录名称：Proceedings of the 18th Chinese Annual Conference on Management Science
- 语种：中文
- 分类号：F323.7;TP18
- 学会代码：ZHYJ
- 会议名称：第十八届中国管理科学学术年会
- 会议地点：中国陕西西安
- 主办单位：中国优选法统筹法与经济数学研究会、西安交通大学、中国科学院科技战略咨询研究院、《中国管理科学》编辑部
- 学会名称：中国优选法统筹法与经济数学研究会
- 页数：7
- 文件大小：549k
- 原文格式：O
- 会议级别：全国
摘要
国内猪肉市场价格具有波动大、非线性、非平稳,且样本量少的特点,很难进行预测。为了提高预测精度,并有效解释价格波动的内在经济含义,基于集成预测思想,提出EMD-SVM集成预测模型。首先用经验模态分解方法(EMD)把猪肉市场月度价格分解成若干个不同尺度的,相对平稳的本征模态分量(IMF),按照频率高低,将各IMF分量集成为高频部分、低频部分和残余项三大模块,解决波动大、非平稳问题。在此基础上运用支持向量机(SVM)对3个集成模块分别进行预测,从而解决非线性问题。为了使预测模型最优,SVM的参数用遗传算法进行寻优。最后对3个集成模块的预测结果再次进行集成,重构出猪肉市场价格预测值。为了验证模型的有效性,将EMD-SVM集成预测模型与SVM、EMD-BP、BP的预测结果进行分类比较,其RMSE、MAPE和方向性都明显提高。
The price of pork in domestic market is difficult to predict because of widely fluctuation,nonlinearity non-stationary data and lack of sample.The integrated model EMD-SVM,stands for Empirical Mode Decomposition-Support Vector Machine,is proposed in order to improve the accuracy and interpret economic internal connotation of fluctuation effectively.First of all,the pork market price is decomposed to several different scales,relatively stable the intrinsic mode components(IMF) based on empirical mode decomposition(EMD) method.These IMFs will be divided into three modules such as high frequency part,low frequency part and residual.Thus the problem of fluctuation and non-stationary is settled.The SVM method will be used to calculate the three components so that nonlinearity is solved.For the purpose of optimization,the genetic algorithm(GA) is utilized to analyze the parameters of SVM.Finally,the prediction of market price will be reconstructed by integrating the result from previous three modules again.With the intention of examining the validity,comparing the consequence of SVM,EMD-BP and BP,the RMSE,MAPE and directionality are enhanced obviously.
The price of pork in domestic market is difficult to predict because of widely fluctuation,nonlinearity non-stationary data and lack of sample.The integrated model EMD-SVM,stands for Empirical Mode Decomposition-Support Vector Machine,is proposed in order to improve the accuracy and interpret economic internal connotation of fluctuation effectively.First of all,the pork market price is decomposed to several different scales,relatively stable the intrinsic mode components(IMF) based on empirical mode decomposition(EMD) method.These IMFs will be divided into three modules such as high frequency part,low frequency part and residual.Thus the problem of fluctuation and non-stationary is settled.The SVM method will be used to calculate the three components so that nonlinearity is solved.For the purpose of optimization,the genetic algorithm(GA) is utilized to analyze the parameters of SVM.Finally,the prediction of market price will be reconstructed by integrating the result from previous three modules again.With the intention of examining the validity,comparing the consequence of SVM,EMD-BP and BP,the RMSE,MAPE and directionality are enhanced obviously.
引文
[1]马孝斌,王婷,董霞,等.向量自回归法在生猪价格预测中的应用[J].中国畜牧杂志,2007,(23):4-6.
[2]马雄威,朱再青.灰色神经网络模型在猪肉价格预测中的应用[J].内蒙古农业大学学报,2008,(4):91-93.
[3]Ding Linlin,Meng Jun,Yang Zhenhua.An early warning system of pork price in China based on decision tree[C]International Conference on E-Product E-Service and EEntertainment.7-9 November,2010.Henan,China.
[4]Wang Shouyang,Yu Lean,Lai K K.Crude oil price forecasting with TEI~@I methodology[J].Journal of Systems Science&Complexity,2005,18(2):145-166.
[5]汪寿阳,余乐安,黎健强.TEI~@I方法论及其在外汇汇率预测中的应用[J].管理学报,2007,4(1):21-27.
[6]Xue Xiaoming,Zhou Jianzhong,Xu Yanhe.An adaptively fast ensemble empirical mode decomposition method and its applications to rolling element bearing fault diagnosis[J].Mechanical Systems and Signal Processing,2015(62):444-459.
[7]Huang N E,Shen Zheng,Long S R,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proceedings of the Royal Society A Mathematical Physical&Engineering Sciences,1998,454(1971):903-995.
[8]Zhang Bo,Wang Jun.Fang Wen.Volatility behavior of visibility graph EMD financial time series from Ising interacting system[J].Physica A.2015(432):301-314.
[9]刘慧婷,倪志伟,李建洋.经验模态分解方法及其实现[J].计算机工程与应用.2006,42(32):44-47.
[10]孙晓蕾,姚晓阳,杨玉英等.国家风险动态性的多尺度特征提取与识别:以OPEC国家为例[J].中国管理科学,2015,23(4):1-10.
[11]范琳琳,王红瑞等.基于T检验的水文时间序列HTT分析方法及应用[J].系统工程理论与实践.2015,35(5):1324-1331.
[12]Vapnik V.The nature of statistical leaning theory,[M].New York:Springer Verlag,1999.
[13]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报.2011(40):2-10.
[14]Holland J H.Adaptation in natural and artificial systems[M].Ann Arbor:University of Michigan press,1975.
[15]郑春红,焦李成,丁爱玲.基于启发式遗传算法的SVM模型自动选择[J].控制理论与应用,2006,23(2):187-192.
[16]Zhang Xu,Lai K K,Wang Shouyang.A new approach for crude oil price analysis based on empirical mode decomposition[J].Energy Economics,2008,30(3):905-918.
[17]杨云飞,鲍玉昆,胡忠义,等.基于EMD和SVMs的原油价格预测方法[J].管理学报,2010,7(12):1884-1889.
[18]任涵均.基于EMD的中国大豆收购价格的预测[J].金融经济.2015,(10):60-63.
[2]马雄威,朱再青.灰色神经网络模型在猪肉价格预测中的应用[J].内蒙古农业大学学报,2008,(4):91-93.
[3]Ding Linlin,Meng Jun,Yang Zhenhua.An early warning system of pork price in China based on decision tree[C]International Conference on E-Product E-Service and EEntertainment.7-9 November,2010.Henan,China.
[4]Wang Shouyang,Yu Lean,Lai K K.Crude oil price forecasting with TEI~@I methodology[J].Journal of Systems Science&Complexity,2005,18(2):145-166.
[5]汪寿阳,余乐安,黎健强.TEI~@I方法论及其在外汇汇率预测中的应用[J].管理学报,2007,4(1):21-27.
[6]Xue Xiaoming,Zhou Jianzhong,Xu Yanhe.An adaptively fast ensemble empirical mode decomposition method and its applications to rolling element bearing fault diagnosis[J].Mechanical Systems and Signal Processing,2015(62):444-459.
[7]Huang N E,Shen Zheng,Long S R,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proceedings of the Royal Society A Mathematical Physical&Engineering Sciences,1998,454(1971):903-995.
[8]Zhang Bo,Wang Jun.Fang Wen.Volatility behavior of visibility graph EMD financial time series from Ising interacting system[J].Physica A.2015(432):301-314.
[9]刘慧婷,倪志伟,李建洋.经验模态分解方法及其实现[J].计算机工程与应用.2006,42(32):44-47.
[10]孙晓蕾,姚晓阳,杨玉英等.国家风险动态性的多尺度特征提取与识别:以OPEC国家为例[J].中国管理科学,2015,23(4):1-10.
[11]范琳琳,王红瑞等.基于T检验的水文时间序列HTT分析方法及应用[J].系统工程理论与实践.2015,35(5):1324-1331.
[12]Vapnik V.The nature of statistical leaning theory,[M].New York:Springer Verlag,1999.
[13]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报.2011(40):2-10.
[14]Holland J H.Adaptation in natural and artificial systems[M].Ann Arbor:University of Michigan press,1975.
[15]郑春红,焦李成,丁爱玲.基于启发式遗传算法的SVM模型自动选择[J].控制理论与应用,2006,23(2):187-192.
[16]Zhang Xu,Lai K K,Wang Shouyang.A new approach for crude oil price analysis based on empirical mode decomposition[J].Energy Economics,2008,30(3):905-918.
[17]杨云飞,鲍玉昆,胡忠义,等.基于EMD和SVMs的原油价格预测方法[J].管理学报,2010,7(12):1884-1889.
[18]任涵均.基于EMD的中国大豆收购价格的预测[J].金融经济.2015,(10):60-63.