基于数据挖掘的股价走势预测
|
摘要
随着社会经济的发展和人们投资意识的不断增强,股票已经成为投资理财的一种重要工具,从而股票走势的预测具有十分重要的意义,然而,股票市场是一个极其复杂的系统,股价走势的预测问题是一个非常困难的问题,尽管如此,股价走势的预测还是引起了越来越多人的关注和研究。
数据挖掘,是90年代中后期发展起来的人工智能分支,它以发现海量数据中隐含的、新颖的、有价值的信息和模式为目标,是一种高层次的数据分析。股票市场中积累了大量的交易数据,数据中隐含了大量有用的信息,采用数据挖掘的相关技术对股市数据进行分析,探索股价走势中的规律,建立股价走势的预测模型,无疑具有重大的现实意义。
本文以中国股票市场为背景,利用数具挖掘的相关技术建立了用于预测股票走势的定性预测模型和定量预测模型,并得到了比较好的结果,由于采用的数据是沪深股市其中550只个股近十年累计约120万个交易日的数据,数据具有很好的代表性,因此,模型具有良好的泛化能力,模型产生的结论也具有较强的说服力,模型具有一定的参考价值,同时,本文的研究页表明了采用数据挖掘的相关技术进行股价预测是可行的。
With social-economic development and strengthening of people’s investment consciousness, The forecast of stock price movement is very important and meaningful,stocks have become an important investment instrument. However, as stock market is a very complicated system, forecast of stock price movement has become very difficult; despite the challenges, people are paying more attention to this area and are doing more research.
Data-mining ,a new Artificial Intelligence branch developed since1990s',focus on discovering valuable modes which are hidden in mega-data and it is high-level of data analysis. a large number of stock market transaction data is accumulated , a great deal of useful information is implied in stock data,Thus, using data-mining techniques to analyze stock data and explore the law in the stock price movement,building forecast model on stock price movement is very meaningful.
This paper builds both the qualitative and quantitative forecast models of stock price movement using data-mining on the Chinese stock market with satisfactory results. Among the stocks traded on the Shenzhen and Shanghai Stock Exchange, 550 stocks have records of 1.2 million trading days accumulatively, Data are well represented, so the models can be widely extended with pervasive conclusions. We believe it is workable to forecast stock price movement using data-mining.
数据挖掘,是90年代中后期发展起来的人工智能分支,它以发现海量数据中隐含的、新颖的、有价值的信息和模式为目标,是一种高层次的数据分析。股票市场中积累了大量的交易数据,数据中隐含了大量有用的信息,采用数据挖掘的相关技术对股市数据进行分析,探索股价走势中的规律,建立股价走势的预测模型,无疑具有重大的现实意义。
本文以中国股票市场为背景,利用数具挖掘的相关技术建立了用于预测股票走势的定性预测模型和定量预测模型,并得到了比较好的结果,由于采用的数据是沪深股市其中550只个股近十年累计约120万个交易日的数据,数据具有很好的代表性,因此,模型具有良好的泛化能力,模型产生的结论也具有较强的说服力,模型具有一定的参考价值,同时,本文的研究页表明了采用数据挖掘的相关技术进行股价预测是可行的。
With social-economic development and strengthening of people’s investment consciousness, The forecast of stock price movement is very important and meaningful,stocks have become an important investment instrument. However, as stock market is a very complicated system, forecast of stock price movement has become very difficult; despite the challenges, people are paying more attention to this area and are doing more research.
Data-mining ,a new Artificial Intelligence branch developed since1990s',focus on discovering valuable modes which are hidden in mega-data and it is high-level of data analysis. a large number of stock market transaction data is accumulated , a great deal of useful information is implied in stock data,Thus, using data-mining techniques to analyze stock data and explore the law in the stock price movement,building forecast model on stock price movement is very meaningful.
This paper builds both the qualitative and quantitative forecast models of stock price movement using data-mining on the Chinese stock market with satisfactory results. Among the stocks traded on the Shenzhen and Shanghai Stock Exchange, 550 stocks have records of 1.2 million trading days accumulatively, Data are well represented, so the models can be widely extended with pervasive conclusions. We believe it is workable to forecast stock price movement using data-mining.
引文
[1] Mendelsohn Stein. Fundamental analysis meets the neural etwork Futures,1991,20(10):22-24P
[2] Sheng-chai Ch ,Hung-pin Chen., Chun-hao Cheng. A prediction Approach For stock index future using Grey Model and Neural Networks[c] international joint conference on Neural Nwtworks ,1999:3850-3855P
[3]霍建军.基于灰色系统和神经网络的股价指数预测:[硕士学位论文].哈尔滨工程大学,2006.
[4]张秀艳,徐立本.基于神经网络集成系统的股市预测模型.系统工程理论与实践,2003,9:67~70
[5]张丽娜.支持向量机对股市的预测及实证分析:[硕士学位论文].青岛大学,2007.
[6]孟慧慧.一种基于组合技术的股票预测方法的研究:[硕士学位论文].青岛理工大学,2007
[7]中国证券业协会编.证券投资分析.北京:中国财经经济出版社,2007
[8]中国证券业协会编.证券市场基础知识.北京:中国财经经济出版社,2007
[9]苑莹,庄新田.股票市场非有效性与可预测性的实证研究.统计与决策,2007,4:7-8.
[10]常松,何健敏.匹配追踪算法对上海股市价格行为的分析.管理工程学报,2002,16(1):88-90.
[11]马树才,赵丰义.中国沪深股市可预测性研究.当代经济管理,2007,9(3):103-106.
[12]潘琳.基于小波分析与神经网络的股票市场预测应用研究:[硕士学位论文].武汉理工大,2006.
[13]孙士保,泰克云.改进的K-平均聚类算法研究.计算机工程,2007.33(13)
[14] JiaWei Han ,Micheline Kamber.数据挖掘概念与技术.范明,孟小峰译.北京:机械工业出版社,2007
[15]樊玫.基于kohonen神经网络的用户访问模式挖掘模型的研究:[硕士学位论文].南昌大学,2007
[16]梁盾.数据挖掘算法与应用.北京:北京大学出版社,2006
[17] Pang-Ning Tan,Michael Steinbach,,Vipin Kumar.数据挖掘导论.范宏建,范明译.北京:人民邮电出版社,2006
[18]刘赣州.资本市场与资本配置效率:基于中国的实证分析.当代经济研究,2003,11 :69~2.
[19]李国欣,郭彩琴,郭璐.中国股票市场的效率分析.统计与决策,2005,08X :90~92
[20]曹筠.采用小波变换建立价格波动幅度的预测模型.商场现代化,2005,5:2~3.
[21]李国平,林敬松.一种改进的灰色模型在股票价格预测中的应用.商场现代化,2005,05X:188~189.
[22] Mac Queen J. Some Methods for Classification and Analysis of Multivariate Observations[C]//Proc. of the 5th Berkeley Symposium on Mathematical Statistics and Probability,1967: 281~297.
[23] Kaufman J, Rousseeuw P J. Finding Groups in Data: An Introduction to Cluster Analysis . New York: John Wiley & Sons, 1990.
[24] Ester M, Kriegel H P, Sander J, et al. A Density-based Algorithm for Discovering Clusters in Large Spatial Databases[C]//Proc. of 1996 Intl. Conf. on Knowledge Discovery and Data Mining, Portland, OR,1996-08: 226~231.
[25] Wang W, Yang J, Muntz R. STING: A Statistical Information Grid Approach to Spatial Data Mining[C]//Proc. of 1997 Intl. Conf. on Very Large Databases, Athens, Greece,1997-08: 186~195.
[26] Han Jiawei, Kamber M. Data Mining: Concepts and Techniques[M].San Francisco: Morgan Kaufmann Publishers,2000.
[27] Kohonen T. Self-organized Formation of Topologically Correct Feature Maps[J]. Biological Cybernetics, 1982,43(1): 59~69.
[28]谢邦昌.数据挖掘Clementine应用实务.北京:机械工业出版社,2008.
[29] Schierholt .Karsten. Stock Market Prediction Using Different Neural Network Classification Architectures. IEEE/IAFE conference on Computational intelligence for Financial Engineering ,1999,(1):72~78
[30]昊贻鼎,朱翔等.基于神经网络的证券市场预测.计算机应用,2002,22 (5):32-33
[31]张文彤.SPSS统计分析高级教程.北京:高等教育出版社,2004
[32]李宗福,邓琼波,李桓.Kohonen .SOFM神经网络及其演化研究.计算机工程与设计,2004,25(10):1729~1830.
[33]杨俊凯.基于长期记忆特征时序分析的中国股市效率性检验.西安金融,2005,1:37~38
[34]蒋艳凰,杨学军.多层组合分类器研究.计算机工程与科学,2004,26(6):67~69,76.
[35]叶德谦,金大兵,杨樱.基于强化学习的股票预测系统的研究与设计.微计算机信息,2006,22(23):149~151.
[36]李国平.中国股票市场的可预测性研究.高职论丛,2006,9:20~21
[37]张敏,刘凤根.股票价格的可预测性与市场有效性.湖南商学院学报,2007,14(5):70~73.
[38]李正学.神经网络理论及其在股市短期预侧中的应用:[博士学位论文].吉林大学,2000.
[39]邓乃杨,田英杰.数据挖掘中的新方法—支持向量机.北京:科学出版社,2004.
[40]冯冬青.基于神经网网络的价值预测方法.计算机工程,2005,31(6): 160
[41] WISHARTD.K~means clustering with outlier detection: the 25th Annual Conference of the German Classification Society[C].Munich:University of Munich.2001: 14~16.
[42]飞科技产品研究中心.神经网络理论与MATLAB7实现.北京:电子工业出版社,2005
[43] J.A.Hartigan.Clustering Algorithms.Wiley.New York.,1975
[44] B.Lan.S.Bressan.,B.C.Ool.Making web servers Pushier.Lecture Notes in Artificial Intelligence.vol.1836.Springer~Verlag.Berlin,2000,pp.112~125
[2] Sheng-chai Ch ,Hung-pin Chen., Chun-hao Cheng. A prediction Approach For stock index future using Grey Model and Neural Networks[c] international joint conference on Neural Nwtworks ,1999:3850-3855P
[3]霍建军.基于灰色系统和神经网络的股价指数预测:[硕士学位论文].哈尔滨工程大学,2006.
[4]张秀艳,徐立本.基于神经网络集成系统的股市预测模型.系统工程理论与实践,2003,9:67~70
[5]张丽娜.支持向量机对股市的预测及实证分析:[硕士学位论文].青岛大学,2007.
[6]孟慧慧.一种基于组合技术的股票预测方法的研究:[硕士学位论文].青岛理工大学,2007
[7]中国证券业协会编.证券投资分析.北京:中国财经经济出版社,2007
[8]中国证券业协会编.证券市场基础知识.北京:中国财经经济出版社,2007
[9]苑莹,庄新田.股票市场非有效性与可预测性的实证研究.统计与决策,2007,4:7-8.
[10]常松,何健敏.匹配追踪算法对上海股市价格行为的分析.管理工程学报,2002,16(1):88-90.
[11]马树才,赵丰义.中国沪深股市可预测性研究.当代经济管理,2007,9(3):103-106.
[12]潘琳.基于小波分析与神经网络的股票市场预测应用研究:[硕士学位论文].武汉理工大,2006.
[13]孙士保,泰克云.改进的K-平均聚类算法研究.计算机工程,2007.33(13)
[14] JiaWei Han ,Micheline Kamber.数据挖掘概念与技术.范明,孟小峰译.北京:机械工业出版社,2007
[15]樊玫.基于kohonen神经网络的用户访问模式挖掘模型的研究:[硕士学位论文].南昌大学,2007
[16]梁盾.数据挖掘算法与应用.北京:北京大学出版社,2006
[17] Pang-Ning Tan,Michael Steinbach,,Vipin Kumar.数据挖掘导论.范宏建,范明译.北京:人民邮电出版社,2006
[18]刘赣州.资本市场与资本配置效率:基于中国的实证分析.当代经济研究,2003,11 :69~2.
[19]李国欣,郭彩琴,郭璐.中国股票市场的效率分析.统计与决策,2005,08X :90~92
[20]曹筠.采用小波变换建立价格波动幅度的预测模型.商场现代化,2005,5:2~3.
[21]李国平,林敬松.一种改进的灰色模型在股票价格预测中的应用.商场现代化,2005,05X:188~189.
[22] Mac Queen J. Some Methods for Classification and Analysis of Multivariate Observations[C]//Proc. of the 5th Berkeley Symposium on Mathematical Statistics and Probability,1967: 281~297.
[23] Kaufman J, Rousseeuw P J. Finding Groups in Data: An Introduction to Cluster Analysis . New York: John Wiley & Sons, 1990.
[24] Ester M, Kriegel H P, Sander J, et al. A Density-based Algorithm for Discovering Clusters in Large Spatial Databases[C]//Proc. of 1996 Intl. Conf. on Knowledge Discovery and Data Mining, Portland, OR,1996-08: 226~231.
[25] Wang W, Yang J, Muntz R. STING: A Statistical Information Grid Approach to Spatial Data Mining[C]//Proc. of 1997 Intl. Conf. on Very Large Databases, Athens, Greece,1997-08: 186~195.
[26] Han Jiawei, Kamber M. Data Mining: Concepts and Techniques[M].San Francisco: Morgan Kaufmann Publishers,2000.
[27] Kohonen T. Self-organized Formation of Topologically Correct Feature Maps[J]. Biological Cybernetics, 1982,43(1): 59~69.
[28]谢邦昌.数据挖掘Clementine应用实务.北京:机械工业出版社,2008.
[29] Schierholt .Karsten. Stock Market Prediction Using Different Neural Network Classification Architectures. IEEE/IAFE conference on Computational intelligence for Financial Engineering ,1999,(1):72~78
[30]昊贻鼎,朱翔等.基于神经网络的证券市场预测.计算机应用,2002,22 (5):32-33
[31]张文彤.SPSS统计分析高级教程.北京:高等教育出版社,2004
[32]李宗福,邓琼波,李桓.Kohonen .SOFM神经网络及其演化研究.计算机工程与设计,2004,25(10):1729~1830.
[33]杨俊凯.基于长期记忆特征时序分析的中国股市效率性检验.西安金融,2005,1:37~38
[34]蒋艳凰,杨学军.多层组合分类器研究.计算机工程与科学,2004,26(6):67~69,76.
[35]叶德谦,金大兵,杨樱.基于强化学习的股票预测系统的研究与设计.微计算机信息,2006,22(23):149~151.
[36]李国平.中国股票市场的可预测性研究.高职论丛,2006,9:20~21
[37]张敏,刘凤根.股票价格的可预测性与市场有效性.湖南商学院学报,2007,14(5):70~73.
[38]李正学.神经网络理论及其在股市短期预侧中的应用:[博士学位论文].吉林大学,2000.
[39]邓乃杨,田英杰.数据挖掘中的新方法—支持向量机.北京:科学出版社,2004.
[40]冯冬青.基于神经网网络的价值预测方法.计算机工程,2005,31(6): 160
[41] WISHARTD.K~means clustering with outlier detection: the 25th Annual Conference of the German Classification Society[C].Munich:University of Munich.2001: 14~16.
[42]飞科技产品研究中心.神经网络理论与MATLAB7实现.北京:电子工业出版社,2005
[43] J.A.Hartigan.Clustering Algorithms.Wiley.New York.,1975
[44] B.Lan.S.Bressan.,B.C.Ool.Making web servers Pushier.Lecture Notes in Artificial Intelligence.vol.1836.Springer~Verlag.Berlin,2000,pp.112~125