基于最小二乘小波支持向量机的股票期货市场预测
|
摘要
支持向量机预测理论是九十年代后期发展起来的新的数据挖掘技术,它是建立在统计学习理论基础上的通用学习方法,着重于研究小样本条件下的预测规律。由于股票期货市场是复杂的非线性系统,传统的时间序列预测技术很难揭示其内在的规律,为了更好的对股票期货市场的价格规律进行分析,本文在支持向量机与小波变换的理论基础上推导出了一种新的方法来对股票期货价格的进行预测,并通过仿真实验,分析了其优缺点.本文主要在如下方面进行了研究和探讨:
1.首先论述了时间序列预测的研究状况及方法的优点与不足,然后介绍支持向量机预测、小波理论与相空间重构的相关概念和模型,这些构成了本文的理论基础。
2.构建了小波核的支持向量机模型,并根据核函数的条件,提出了核函数的构造方法,证明了几种小波函数作为核函数的可行性,为发现更多核函数提供了依据。同时分析了核参数的作用及参数寻优的方法。
3.结合最小二乘法,提出了基于小波核的最小二乘支持向量机预测模型,并把该模型应用于沪深300指数与美国原油期货指数的预测,并与标准最小二乘支持向量机及神经网络模型的预测效果进行比较,验证了该方法的优越性,最后分析了该方法的优缺点及进一步的研究方向。
The forecast using Support Vector Machine(SVM) is new data mining technique developed from the 1990s.SVM based on the foundations of Statistical Learning Theory,which is a small-sample statistics and concerns mainly the forecast rule when samples are limited. On account of stock and futures is the complicated nonline system,the conventional forecast skill hardly find out the inherent rule.For the sake of furthermore analyze the rule of stock and futures market,the paper deduce a new means to dope out the market base on SVM and wavelt transform,and analyze the advantage and shortcoming pass experiment. The paper will investigate and discuss in as follows:
1.First,research status and the advantage ang shortcoming of method was discussed.And introduce the SVM and wavelt theory and correlation model,which form the textual theoretics basis.
2.Building the model of support vector machine using the wavelt kernel,and basing the conditiong of kernel function proved the feasibility of some wavelt kernel function.The paper brought forward the method of construct kernel function,which offer the thereunder for finding out more kernel function.At the same time,analyze the effect and excellent parameter of kernel function.
3.We brought forward the forecasting model of Least squares wavelt support vector machines,which is applied in Hushen 300 index and America crude oil futures index.we compare on the forecast effect of standard LSSVM and ANN model,approving the advantage of the model in forecasting.At last,we analysis the model advantage and shortcoming and the further study aspect.
1.首先论述了时间序列预测的研究状况及方法的优点与不足,然后介绍支持向量机预测、小波理论与相空间重构的相关概念和模型,这些构成了本文的理论基础。
2.构建了小波核的支持向量机模型,并根据核函数的条件,提出了核函数的构造方法,证明了几种小波函数作为核函数的可行性,为发现更多核函数提供了依据。同时分析了核参数的作用及参数寻优的方法。
3.结合最小二乘法,提出了基于小波核的最小二乘支持向量机预测模型,并把该模型应用于沪深300指数与美国原油期货指数的预测,并与标准最小二乘支持向量机及神经网络模型的预测效果进行比较,验证了该方法的优越性,最后分析了该方法的优缺点及进一步的研究方向。
The forecast using Support Vector Machine(SVM) is new data mining technique developed from the 1990s.SVM based on the foundations of Statistical Learning Theory,which is a small-sample statistics and concerns mainly the forecast rule when samples are limited. On account of stock and futures is the complicated nonline system,the conventional forecast skill hardly find out the inherent rule.For the sake of furthermore analyze the rule of stock and futures market,the paper deduce a new means to dope out the market base on SVM and wavelt transform,and analyze the advantage and shortcoming pass experiment. The paper will investigate and discuss in as follows:
1.First,research status and the advantage ang shortcoming of method was discussed.And introduce the SVM and wavelt theory and correlation model,which form the textual theoretics basis.
2.Building the model of support vector machine using the wavelt kernel,and basing the conditiong of kernel function proved the feasibility of some wavelt kernel function.The paper brought forward the method of construct kernel function,which offer the thereunder for finding out more kernel function.At the same time,analyze the effect and excellent parameter of kernel function.
3.We brought forward the forecasting model of Least squares wavelt support vector machines,which is applied in Hushen 300 index and America crude oil futures index.we compare on the forecast effect of standard LSSVM and ANN model,approving the advantage of the model in forecasting.At last,we analysis the model advantage and shortcoming and the further study aspect.
引文
[1]杨一文,刘贵忠,张宗平.基于嵌入理论和神经网络技术的混沌数据预测及其在股票市场中的应用[J].系统工程理论与实践,2001,6(6):52-58
[2]曲文龙,樊广铨,杨炳儒.基于支持向量机的复杂时间序列预测研究[J].计算机工程,2005,31(23):2513-2522
[3]Vapnik V N.Statistical Learning Theory[M].New York:Wiley,1998
[4]张学工译.关于统计学习理论与支持向量机[J].自动化学报,2000,4(26):32-420
[5](美)马丁·J.普林格.技术分析[M].中国金融出版社
[6]赵振全,蒋瑛馄,代丽,卢世春.我国证券市场结构分析及优化[J].数量经济技术经济研究,2001(6):36-38
[7]胡广生,臧保将.股市时间序列的非线性分析[J].北京交通大学学报,2006,30(6)60-65
[8]埃德加·E.彼得斯,王小东译.《资本市场的混饨与秩序》[M].经济科学出版社,1999
[9]Cordon T.J.Integration of forecasting methods,and the frontics of futures research.United Nations University,1994,4(38);323-331
[10]Cordon T.J.,Greenspan.D.Chaos and fractals:new tool for technological and forecasting social.Technological Forecasting & Social Change,1988,1(34):1-25
[11]吕金虎,陆君安,陈士华.混沌时间序列分析及其应用[M].武汉:武汉大学出版社
[12]姚洪兴,姜桂仁,耿霞.相空间重构中参数确定方法的新探讨[J].江苏大学学报,2005,12(26-6):82-85
[13]姚洪兴,石桃丽.一类金融市场模型的混沌控制.微计算机信息,2007,23(9):50-52
[14]卢山,王海燕.证券市场的非线性和确定性检验[J].系统工程理论方法应用,2005,14(3):235-238
[15]向小东.小波神经网络预测方法在石油期货价格预测中的应用研究[J].技术经济2006,6(25-6):121-124
[16]姚洪兴,盛昭翰,陈洪香.股市预测中的小波神经网络方法[J].系统工程理论与实践,2002,33(6):33-39
[17]姚洪兴,吴越.多变量经济混沌时序的小波神经网络预测.江苏大学学报(自然科学版),2004,25(4):319-323
[18]Vapnik V.The Nature of Statistical LearningTheory[M].NewYork:Springer Verlag,1995
[19]Mukherjee S,Osuna E,Girosi F.Nonlinearprediction of chaotic time series using supportvector machines[C].In:Proceedings of the IEEE Signal Processing Society Workshop,1997,511-520
[20]Fan A,Palaniswami M.Selecting Bankruptcy Predictors Using a Support Vector Machine Approach[C].In:Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks,2000,6:354-359
[21]Tony V G,Johan A K S,Dirk-Emma B,Annemie L,Gert L,Bruno V,Bart D M,Joos V.Financial time series prediction using leastsquares support vectormachines within theevidence framework(J).IEEE Transactions on Neural Networks,2001,12:809-821
[22]Tay F E H,Cao L J.Modified support vector machines in financial time series forecasting [J].Neurocomputing,2002,48:847-861
[23]Cao L J.Support vector machines experts for time series forecasting[J].Neurocomputing,2003,51(4):321-339
[24]Thissen U,Brakel R V,Weijer A P,Melssen W J,Buydens L M C.Using support vector machines for time series prediction[J].Chemometrics and Intelligent Laboratory Systems,2003,69:35-49
[25]Zhang Z Y,Shi C,Zhang S L,Shi Z Z.Stock Time Series Forecasting Using Support Vector Machines Employing Analyst Recommendations[C].In:Lecture Notes in Computer Science,2006,452-457
[26]祝金荣.基于支持向量机的石油期货价格预测[J].工业技术经济.2007,26(2):58-61
[27]Wei Huang,Yoshiteru Nakarnori,Shou-Yang Wang,Forecasting stock market movement direction with support vector machine[J],Computers&Operations Research,2005(32);2513-2522
[28]L.J.Cao,S.S.Keerthi,C.J.Ong,P.Uvaraj,X.J.Fu,H.P.Developing parallel sequential minimal optimization for fast training support vector machine[J].Lee Neurocomputing,2006:93-104
[29]曹丽娟,王小明.金融工程的支持向量机方法[M],上海:上海财经大学出版社,2007
[30]李国正,王猛,曾华军.支持向量机导论[M],北京:电子工业出版社,2004
[31]邓乃杨,田英杰.数据挖掘中的新方法一支持向量机[M].北京:科学出版社
[32](美)Vladimir N Vapnik,张学工译.统计学习理论的本质[M].北京:清华大学出版社,2000
[33]Cherkassky V,Mulier F.Vapnik-Chervonenkis(VC) Learning Theory and Its Applications[J].IEEE Transactions on Neural Networks,1999,10:985-987
[34]Vapnik V,Levin E,Le Cun Y.Measuring the VC-dimension of a learning machine[J].Neural Computation,1994,(6):851-876
[35]李忠浩.支持向量机算法及其应用研究(硕士学位论文).大连理工大学,2006
[36]冯象初,甘小冰,宋国乡.数值泛函与小波理论[M].西安:西安电子科技大学出版社,2003
[37]张国云.支持向量机算法及其应用研究(博士学位论文).湖南大学,2006
[38]ShuyuanYang,MinWang,LichengJiao.Ridgeletkemelregression[J].Neurocomputing,2007,(7):3046-3055
[39]Senjian An,Wanquan Liu,Svetha Venkatesh.Fast cross-validation algorithms squares support vector machine and kernel ridge regression[J].Pattern Recognition,40(2007):2154-2162
[40]郭家芳.基于小波分析和支持向量机的股票指数预测(硕士学位论文).武汉理工大学,2006
[41]范玉刚,李平,宋执环.动态加权最小二乘支持向量机[J].控制与决策,2006,21(10):1129-1133
[42]杨延西,刘丁,李琦.小波及混沌学习神经网络在短期电力负荷预测中的应用[J].计算机编程与应用.2003,39(23):217-220
[43]江田汉,束炯.基于的混沌时间序列的多步预测[J].控制与决策,2006,21(1),77-80
[44]李水根,吴纪桃.分形与小波[M].北京:科学出版社,2002
[45]Michael E,Matheny Frederic S,Resnic Nipun Arora,Lucila Ohno-Machado.Effects of SVM parameter optimization on discriminationand calibration for post-procedural PCI mortality[J].Journal of Biomedical Informatics,2007,6(40):688-697
[2]曲文龙,樊广铨,杨炳儒.基于支持向量机的复杂时间序列预测研究[J].计算机工程,2005,31(23):2513-2522
[3]Vapnik V N.Statistical Learning Theory[M].New York:Wiley,1998
[4]张学工译.关于统计学习理论与支持向量机[J].自动化学报,2000,4(26):32-420
[5](美)马丁·J.普林格.技术分析[M].中国金融出版社
[6]赵振全,蒋瑛馄,代丽,卢世春.我国证券市场结构分析及优化[J].数量经济技术经济研究,2001(6):36-38
[7]胡广生,臧保将.股市时间序列的非线性分析[J].北京交通大学学报,2006,30(6)60-65
[8]埃德加·E.彼得斯,王小东译.《资本市场的混饨与秩序》[M].经济科学出版社,1999
[9]Cordon T.J.Integration of forecasting methods,and the frontics of futures research.United Nations University,1994,4(38);323-331
[10]Cordon T.J.,Greenspan.D.Chaos and fractals:new tool for technological and forecasting social.Technological Forecasting & Social Change,1988,1(34):1-25
[11]吕金虎,陆君安,陈士华.混沌时间序列分析及其应用[M].武汉:武汉大学出版社
[12]姚洪兴,姜桂仁,耿霞.相空间重构中参数确定方法的新探讨[J].江苏大学学报,2005,12(26-6):82-85
[13]姚洪兴,石桃丽.一类金融市场模型的混沌控制.微计算机信息,2007,23(9):50-52
[14]卢山,王海燕.证券市场的非线性和确定性检验[J].系统工程理论方法应用,2005,14(3):235-238
[15]向小东.小波神经网络预测方法在石油期货价格预测中的应用研究[J].技术经济2006,6(25-6):121-124
[16]姚洪兴,盛昭翰,陈洪香.股市预测中的小波神经网络方法[J].系统工程理论与实践,2002,33(6):33-39
[17]姚洪兴,吴越.多变量经济混沌时序的小波神经网络预测.江苏大学学报(自然科学版),2004,25(4):319-323
[18]Vapnik V.The Nature of Statistical LearningTheory[M].NewYork:Springer Verlag,1995
[19]Mukherjee S,Osuna E,Girosi F.Nonlinearprediction of chaotic time series using supportvector machines[C].In:Proceedings of the IEEE Signal Processing Society Workshop,1997,511-520
[20]Fan A,Palaniswami M.Selecting Bankruptcy Predictors Using a Support Vector Machine Approach[C].In:Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks,2000,6:354-359
[21]Tony V G,Johan A K S,Dirk-Emma B,Annemie L,Gert L,Bruno V,Bart D M,Joos V.Financial time series prediction using leastsquares support vectormachines within theevidence framework(J).IEEE Transactions on Neural Networks,2001,12:809-821
[22]Tay F E H,Cao L J.Modified support vector machines in financial time series forecasting [J].Neurocomputing,2002,48:847-861
[23]Cao L J.Support vector machines experts for time series forecasting[J].Neurocomputing,2003,51(4):321-339
[24]Thissen U,Brakel R V,Weijer A P,Melssen W J,Buydens L M C.Using support vector machines for time series prediction[J].Chemometrics and Intelligent Laboratory Systems,2003,69:35-49
[25]Zhang Z Y,Shi C,Zhang S L,Shi Z Z.Stock Time Series Forecasting Using Support Vector Machines Employing Analyst Recommendations[C].In:Lecture Notes in Computer Science,2006,452-457
[26]祝金荣.基于支持向量机的石油期货价格预测[J].工业技术经济.2007,26(2):58-61
[27]Wei Huang,Yoshiteru Nakarnori,Shou-Yang Wang,Forecasting stock market movement direction with support vector machine[J],Computers&Operations Research,2005(32);2513-2522
[28]L.J.Cao,S.S.Keerthi,C.J.Ong,P.Uvaraj,X.J.Fu,H.P.Developing parallel sequential minimal optimization for fast training support vector machine[J].Lee Neurocomputing,2006:93-104
[29]曹丽娟,王小明.金融工程的支持向量机方法[M],上海:上海财经大学出版社,2007
[30]李国正,王猛,曾华军.支持向量机导论[M],北京:电子工业出版社,2004
[31]邓乃杨,田英杰.数据挖掘中的新方法一支持向量机[M].北京:科学出版社
[32](美)Vladimir N Vapnik,张学工译.统计学习理论的本质[M].北京:清华大学出版社,2000
[33]Cherkassky V,Mulier F.Vapnik-Chervonenkis(VC) Learning Theory and Its Applications[J].IEEE Transactions on Neural Networks,1999,10:985-987
[34]Vapnik V,Levin E,Le Cun Y.Measuring the VC-dimension of a learning machine[J].Neural Computation,1994,(6):851-876
[35]李忠浩.支持向量机算法及其应用研究(硕士学位论文).大连理工大学,2006
[36]冯象初,甘小冰,宋国乡.数值泛函与小波理论[M].西安:西安电子科技大学出版社,2003
[37]张国云.支持向量机算法及其应用研究(博士学位论文).湖南大学,2006
[38]ShuyuanYang,MinWang,LichengJiao.Ridgeletkemelregression[J].Neurocomputing,2007,(7):3046-3055
[39]Senjian An,Wanquan Liu,Svetha Venkatesh.Fast cross-validation algorithms squares support vector machine and kernel ridge regression[J].Pattern Recognition,40(2007):2154-2162
[40]郭家芳.基于小波分析和支持向量机的股票指数预测(硕士学位论文).武汉理工大学,2006
[41]范玉刚,李平,宋执环.动态加权最小二乘支持向量机[J].控制与决策,2006,21(10):1129-1133
[42]杨延西,刘丁,李琦.小波及混沌学习神经网络在短期电力负荷预测中的应用[J].计算机编程与应用.2003,39(23):217-220
[43]江田汉,束炯.基于的混沌时间序列的多步预测[J].控制与决策,2006,21(1),77-80
[44]李水根,吴纪桃.分形与小波[M].北京:科学出版社,2002
[45]Michael E,Matheny Frederic S,Resnic Nipun Arora,Lucila Ohno-Machado.Effects of SVM parameter optimization on discriminationand calibration for post-procedural PCI mortality[J].Journal of Biomedical Informatics,2007,6(40):688-697
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |