基于SVR的中国股票定价研究
摘要
IPO定价是新股发行业务的核心内容,成功的发行不仅可以为公司筹措资金以扩大经营规模,而且有助于公司改善资本结构和治理结构。然而,IPO定价一直是国际金融界公认的最具迷惑性的难题之一,由于定价不合理的IPO抑价等问题经常出现。国内外学者在对IPO定价的研究中,重点着眼于IPO发行定价方式、IPO抑价产生的原因,以及对IPO抑价水平的测度等方面,而对如何制定一个公正、合理、科学的IPO定价模型的研究的探讨则比较少。讨论IPO定价问题,就是讨论上市公司的估值问题。本文沿着成熟股票定价理论的思路,详细分析新股定价与主要影响因素指标的相关性,并以量化的方式建立适合中国市场实际的股票定价模型,力图为IPO定价研究的后续发展起到抛砖引玉的作用。
具体而言,本文分析了五类传统的IPO定价模型:现金流贴现模型、可比公司估值模型、EVA估值模型、多因素定价模型和期权定价模型经济学原理和模型方法,并对它们的适用范围和局限性进行了比较。然后,自上而下地从宏观经济、市场走势、公司财务三个层面选取相关的公开信息数据,构建了IPO定价影响因素的指标体系。最后,建立了五个股票板块的IPO定价模型,包括一个大样本模型——信息板块,四个小样本模型——采掘业板块、金融服务板块、商业贸易板块和社会服务板块。在每个样本板块之下,分别建立三种定价模型来研究。
实证结果表明:1)剔除奇异样本对改善大样本的误差情况并无明显作用;2)剔除奇异样本对提升小样本的定价精度有显著的效果;3)综合考虑宏观、市场和企业因素的三层指标模型的定价精度优于仅考虑财务因素的模型。
本文最后针对研究的局限性,提出了后续的研究方向。
Initial Public Offering’s (IPO) pricing is the core process of the offering business. A successful offering can raise funds to enlarge the firm’s operation size and help to improve both the capital structure and governance structure. Nevertheless, IPO pricing is always regarded as one of the most confused puzzles in finance across the world. Problems such as IPO underpricing come up from time to time. Academicians focus on pricing mode, reasons and range of IPO underpricing when studying of IPO pricing. Researches of how to establish a fair, proper and scientific IPO pricing modern is rare. Discussing the issue of IPO pricing means discussing the valuation of a firm that is going public.Under mature stock pricing theory, this paper analyze the relationship between IPO pricing and main influencing factors, then establish IPO pricing moderns which are suitable for China market. The main purpose is to offer a reference for the future study on IPO pricing.
In detail, this paper analyzes the economic theories and modeling methods of five conventional IPO pricing models and compares their application fields and limitations. These six models includes cash flow discount model, contrast firm valuation model, economic value-added model, multi-factors pricing model and option pricing model. Then, index system of IPO pricing influencing factors based on public information is built up of three layers from macro economics, market situation and firm finance. Finally, IPO pricing models are established which includes one large sample model of information board and four small sample models consisting of mining board, finance board, commerce board and social service board. Under every board, there are three corresponding models.
The research results show that: 1) effect of deleting exceptional samples is not obvious to improve the pricing errors of large sample; 2) deleting exceptional samples can advance the pricing errors of small sample; 3) the pricing error of models with three layer indices of macroeconomic, market and firm factors are better than models with only finance indices factors.
Finally, based on the limitations of this paper, it points out future research directions.
具体而言,本文分析了五类传统的IPO定价模型:现金流贴现模型、可比公司估值模型、EVA估值模型、多因素定价模型和期权定价模型经济学原理和模型方法,并对它们的适用范围和局限性进行了比较。然后,自上而下地从宏观经济、市场走势、公司财务三个层面选取相关的公开信息数据,构建了IPO定价影响因素的指标体系。最后,建立了五个股票板块的IPO定价模型,包括一个大样本模型——信息板块,四个小样本模型——采掘业板块、金融服务板块、商业贸易板块和社会服务板块。在每个样本板块之下,分别建立三种定价模型来研究。
实证结果表明:1)剔除奇异样本对改善大样本的误差情况并无明显作用;2)剔除奇异样本对提升小样本的定价精度有显著的效果;3)综合考虑宏观、市场和企业因素的三层指标模型的定价精度优于仅考虑财务因素的模型。
本文最后针对研究的局限性,提出了后续的研究方向。
Initial Public Offering’s (IPO) pricing is the core process of the offering business. A successful offering can raise funds to enlarge the firm’s operation size and help to improve both the capital structure and governance structure. Nevertheless, IPO pricing is always regarded as one of the most confused puzzles in finance across the world. Problems such as IPO underpricing come up from time to time. Academicians focus on pricing mode, reasons and range of IPO underpricing when studying of IPO pricing. Researches of how to establish a fair, proper and scientific IPO pricing modern is rare. Discussing the issue of IPO pricing means discussing the valuation of a firm that is going public.Under mature stock pricing theory, this paper analyze the relationship between IPO pricing and main influencing factors, then establish IPO pricing moderns which are suitable for China market. The main purpose is to offer a reference for the future study on IPO pricing.
In detail, this paper analyzes the economic theories and modeling methods of five conventional IPO pricing models and compares their application fields and limitations. These six models includes cash flow discount model, contrast firm valuation model, economic value-added model, multi-factors pricing model and option pricing model. Then, index system of IPO pricing influencing factors based on public information is built up of three layers from macro economics, market situation and firm finance. Finally, IPO pricing models are established which includes one large sample model of information board and four small sample models consisting of mining board, finance board, commerce board and social service board. Under every board, there are three corresponding models.
The research results show that: 1) effect of deleting exceptional samples is not obvious to improve the pricing errors of large sample; 2) deleting exceptional samples can advance the pricing errors of small sample; 3) the pricing error of models with three layer indices of macroeconomic, market and firm factors are better than models with only finance indices factors.
Finally, based on the limitations of this paper, it points out future research directions.
引文
[1] Graham B..证券分析[M].北京:中国人民大学出版社,2009
[2] Sharpe W..证券投资理论与资本市场[M].北京:中国经济出版社,1992
[3] Samuelson P..Proof that properly anticipated prices fluctuate randomly[J]. Industrial Management Review,1965,6:41-49
[4] Fama E..Random Walks in Stock Market Prices[J].Financial Analysts Journal,1965(9):10-16
[5] Sung-II Cho.A model for IPO pricing and contract choice decision[J].Quarterly Review of Economics & Finance,2001(3):347-264
[6] Gompers P.,Ishii J and Metrics A..Corporate Governance And Equity Prices[J].The Quarterly Journal of Economics,2003,118(1):107-155
[7] Lowry M. and Schwert W..Is the IPO pricing process efficient? [J].Journal of Financial Economics,2004,71(1):3-26
[8]陈礼林,彭晗.修正的股息现值模型在新股定价中的应用[J].预测,1999(5):74-75
[9]陈述云.新股定价计量模型研究[J].理论新探,2001(5):8-9
[10]田华,宋耀.新股上网定价发行模型的建立和检验[J].财政理论与实践,2002,116:67-69
[11]孙国茂.上海股票市场IPO定价实证分析[J].上海经济研究,2003(10):26-34
[12]孔玉生,王忠杰.新股上市首日定价的因素分析[J].中国管理科学,2003(3):6-9
[13]张岚.新股上市首日定价及上市后合理定价模型介绍[J].当代财经,2003(4):59-62
[14]王铁锋.时变情况下股票定价模型研究[J].数量经济技术经济研究,2003(8):140-142
[15]张小成.基于模糊优化模型下IPO定价的研究与应用[D].重庆大学硕士学位论文,2005
[16]李庆峰.二元结构环境下的IPO定价模型研究[J].财经问题研究.2006,268(3):44-53
[17]冯涛,王永明.中国证券市场IPO定价模型及其实证研究[J].统计与信息论坛.2009,24(1):58-61
[18] Reilly F.and Hatfield K..Investor experience with new stock issue[J].Financial Analysts Journal 1969,25:73-80
[19] Ibbotson R..Price performance of common stock new issues[J].Journal of Financial Economics 1975,2:235-272
[20] Rock K..Why new issues are underpriced[J].Journal of Financial Economics,1986,15:187-212
[21] Beatty P..and Ritter R..Investment banking,reputation,and the underpricing of initial public offerings[J].Journal of Financial Economics 1986,15,213-232
[22] Baron P.and Holmstrom B..The investment banking contract for new issues under a symmetric information: delegation and the incentive problem[J].Journal of Finance,1980,35,1115-1138
[23] Biais B.,Bossaerts P.and Rochet J..An optimal IPO mechanism[J].Review of Economic Studies,2002,69,117-146.
[24] Vapnik V.,Golowich S. and Smola A..Support vector machine for function approximation,regression estimation,and signal processing[J].Advances in Neural Information Processing Systems,1996,9:281-287
[25]张铃.基于核函数的SVM机与三层前向神经网络的关系[J].计算机学报.2002,25(7):696-700
[26] Cortes C. and Vapnik..Support vector networks[J].Machine Learning,1995,20: 273-297
[27] Osuna E.,Freund R. and Girosi F..An improved training algorithm for support vector machines[A].Neural Networks for Signal ProceedingⅦ-Proceedings of the 1997 IEEE Workshop[C],IEEE,1997:267-285
[28] Platt C..Fast training of support vector machines using sequential minimal optimization[A].Schlkoph B..Advances in Kernel Method-Support Vector Learning[C].Cambridge,MIT Press,1998:185-208
[29] Joachims T. . Making large-scale SVM learning practical[A] . Schlkoph B..Advances in Kernel Method-Support Vector Learning[C].Cambridge,MIT Press,1999:169-184
[30] Suykens K. and Vandewale J. . Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300
[31] Chih-Chung Chang and Chih-Jen Lin.LIBSVM: a library for support vector machines,http://www.csie.ntu.edu.tw/~cjlin/libsvm,2001
[32] Chih-Wei Hsu,Chih-Chung Chang and Chih-Jen Lin.A Practical Guide to Support Vector Classification.http://www.csie.ntu.edu.tw/~cjlin,2003
[33] Mukherjee S.,Osuna E. and Girosi F.Nonlinear prediction of chaotic time series using support vector machine[A].Institute of Electrical and Electronics Engineers.In Proceedings of the IEEE Workshop on Neural Networks for Signal Processing[C].New York Press,1997:511–519
[34] Tay F. and Li-juan Cao.Application of support vector machines in financial time series forecasting[J].Omega the International Journal of Management Science,2001,29:309-317
[35] Tay F. and Li-juan Cao.Modified support vector machines in financial time series forecasting[J].Neurocomputing,2002,48:847-861
[36] Kim Kyoung-jae . Financial time series forecasting using support vector machines[J].Neurocompution,2003,55:307-319
[37] Ping-feng Pai and Chih-sheng Lin.A hybrid ARIMA and support vector machines model in stock price forecasting[J].Omega the International Journal of Management Science,2005(6):497-505
[38] Ping-Feng Pai and Wan-Ru Wei.Predicting movement directions of stock index futures by support vector models with data preprocessing[A] . Industrial Engineering and Engineering Management 2007 IEEE International Conference[C].2007:169-173
[39] Ruud S.Underwriter price support and the IPO under pricing puzzle[J].Journal of Financial Economics,1993,34:135-151
[40] Wun-Hua Chen and Jen-Ying Shih.Comparison of support-vector machines and back propagation neural networks in forecasting the six major Asian stock markets[J].International Journal of Electronic Finance,2006(1):49-67
[41] Kuan-Yu Chen.Evolutionary support vector regression modeling for TaiwanStock Exchange Market Weighted Index forecasting[J].Journal of American Academy of Business,2006(8):241-247
[42] Tinic M.Anatomy of Initial public offerings of common stock[J].Journal of Finance,1988,43:789-822
[43] Sheng-Hsun Hsu,Po-An Hsieh and Ting-Chih Chih.A two-stage architecture for stock price forecasting by integrating self-organizing map and support vector regression[J].Expert Systems with Applications,2009,36(4):7947-7951
[44]孙延风,梁艳春.支持向量机的数据依赖型核函数改进算法[J].吉林大学学报,2003,41(7):329-333
[45]李立辉.基于SVR的金融时间序列预测[J].计算机工程与应用,2005,30:221-224
[46]王彦峰,高风.基于支持向量机的股市预测[J].计算机仿真,2006,23(11):256-258
[47]张玉川,张作泉.支持向量机在股票价格预测中的应用[J].北京交通大学学报,2007,31(6):73-76
[48]隋学深,齐中英.基于多尺度特征和支持向量机的股市趋势预测[J].哈尔滨工业大学学报,2008,10(4):256-258
[49]李蓬宁.基于支持向量机的股市组合预测模型研究[J].经济研究导刊,2009,(22): 38-42
[50]陈圣滔,孙玉秋.统计模型在证券投资分析中的应用[J].技术经济与管理研究,2005,(2):37-38
[51] Sharp W.Capital Asset Prices:A Theory of Market Equilibrium under Conditions of Risks[J].Journal of Finance,1964,19:425-442
[52] Markowitz M.Portfolio Selection[J].Journal of Finance,1952,7:77-91
[53] Chi-Jie Lu,Tian-Shyug Lee and Chiu Chih-Chou.Financial time series forecasting using independent component analysis and support vector regression[J].Decision Support Systems,2009,47(2):115-125
[54]田翔,邓飞其.精确在线支持向量回归在股指预测中的应用[J].计算机工程,2005,31(22):18-20
[55]陈诗一.汇率预测:一个新的非参数支持向量回归方法[J].数量经济技术经济研究,2007(5):143-151
[56] Brennan J.and Franks J..Underpricing,ownership and control in initial public offerings of equity securities in the U.K. [J].Journal of Financial Economics,1997,45:391-413
[57] Mello S.and Parsons E..Going public and the ownership structure of the firm[J].Journal of Financial Economics,1998,49:79-109
[58] Ljungqvist A.,Nanda V.and Singh R..Hot markets,investor sentiment,and IPO pricing[J].Journal of Business,2006,79:166-1702
[59] Dorn D..Does sentiment drive the retail demand for IPOs?[J].Journal of Finance,,2003,12:22-26.
[60] Oehler A.,Rummer M.and Smith P.N..IPO pricing and the relative importance of investor sentiment-evidence from Germany[J].2004,12:15-18.
[61]张继强,周勇,张秉麟.IPO折价、逆向选择与分离均衡假说[J].预测,2003,4:46-61.
[62]周孝华,胡国生,苟思.中国股市IPOs高抑价的噪声分析[J].软科学,2005,19:30-33.
[63]王春峰,赵威,房振明.新股流动性风险及其与IPO抑价的关系[J].统计与决策,2006,24:21-25.
[64]夏国恩,邵培基.改进SVR在金融时间序列预测中的应用[J].金融理论与实践,2008,(11):95-98
[65]夏国恩,曾绍华,金炜东.支持向量回归机在铁路客运量时间序列预测中的应用[J].计算机应用研究,2006,(10):186-188
[66]夏国恩,金炜东,张葛祥.改进SVR及其在铁路客运量预测中的应用[J].西南交通大学学报,2007,42(4):494-498
[67]冯兴杰,魏新,王亚楼.基于支持向量回归的旅客吞吐量预测研究[J].计算机工程,2005,31(14):172-173
[68]李博,吴世农.中国股市新股发行(IPOS)的初始收益率研究[J].南开管理评论,2000,5,31-36
[69]杜莘,梁洪昀,宋逢明.中国A股市场首日收益率研究[J].管理科学学报,2001,4:55-61
[70]王春峰,姚锦.新股价值低估的随机前沿分析[J].系统工程,2002,22:22-23
[71]刘力,王汀汀.不应忽略股票的流通权价值——兼论中国股票市场的二元股权结构问题[J].管理世界,2003,9:46-51
[72]朱南,卓贤.对我国股票首次公开发行抑价的实证研究[J].财经科学,2004,2:32-36
[73]李翔,阴永晟.发行管制变迁下的中国股市IPO首日回报率研究[J].经济科学,2004,3:43-53
[74]刘煌辉,熊鹏.股权分置、政府管制和中国IPO抑价[J].经济研究,2005,5:85-95
[75]曹凤岐,董秀良.我国IPO定价合理性的实证分析[J].财经研究,2006,32:1-3
[76]张锦莉.沪深股市IPO溢价现象的研究[D].暨南大学硕士学位论文,2008
[77]刘路.中国IPO首日超额收益理论与实证分析[D].复旦大学硕士学位论文,2008
[78]唐勇.发售机制、信息偏倚与IPO抑价[D].浙江大学硕士学位论文,2009
[79]李允默.我国新股上市首日溢价研究[D].武汉科技大学硕士学位论文,2009
[80]李冬琴,王丽铮,王呈方.一种新的支持向量回归算法及其在集装箱吞吐量预测中的应用[J].水运工程,2007,(5):16-19
[81]董辉,傅鹤林,冷伍明.支持向量回归的时间序列回归与预测[J].系统仿真学报,2006,18(7):1785-1788
[82]邵信光,杨慧中,石晨曦.ε-不敏感支持向量回归在化工数据建模中的应用[J].东南大学学报(自然科学版),2004,34:215-218
[83]黎展求,朱栋华,刘冬岩.基于支持向量回归和小波包的供热负荷预测[J].暖通空调,2007,37(2):1-5
[84]徐红敏,刘亚禄.改进的支持向量回归机及其在水质预测中的应用[J].统计与决策,2006,(23):12-13
[85]魏宏业,吕永波,何崴.基与数据挖掘的企业合作伙伴的寻求方法研究.管理科学学报,2004,7(1):60-64
[86] Vapnik V..统计学习理论的本质[M].北京:清华大学出版社,2000:11-155
[2] Sharpe W..证券投资理论与资本市场[M].北京:中国经济出版社,1992
[3] Samuelson P..Proof that properly anticipated prices fluctuate randomly[J]. Industrial Management Review,1965,6:41-49
[4] Fama E..Random Walks in Stock Market Prices[J].Financial Analysts Journal,1965(9):10-16
[5] Sung-II Cho.A model for IPO pricing and contract choice decision[J].Quarterly Review of Economics & Finance,2001(3):347-264
[6] Gompers P.,Ishii J and Metrics A..Corporate Governance And Equity Prices[J].The Quarterly Journal of Economics,2003,118(1):107-155
[7] Lowry M. and Schwert W..Is the IPO pricing process efficient? [J].Journal of Financial Economics,2004,71(1):3-26
[8]陈礼林,彭晗.修正的股息现值模型在新股定价中的应用[J].预测,1999(5):74-75
[9]陈述云.新股定价计量模型研究[J].理论新探,2001(5):8-9
[10]田华,宋耀.新股上网定价发行模型的建立和检验[J].财政理论与实践,2002,116:67-69
[11]孙国茂.上海股票市场IPO定价实证分析[J].上海经济研究,2003(10):26-34
[12]孔玉生,王忠杰.新股上市首日定价的因素分析[J].中国管理科学,2003(3):6-9
[13]张岚.新股上市首日定价及上市后合理定价模型介绍[J].当代财经,2003(4):59-62
[14]王铁锋.时变情况下股票定价模型研究[J].数量经济技术经济研究,2003(8):140-142
[15]张小成.基于模糊优化模型下IPO定价的研究与应用[D].重庆大学硕士学位论文,2005
[16]李庆峰.二元结构环境下的IPO定价模型研究[J].财经问题研究.2006,268(3):44-53
[17]冯涛,王永明.中国证券市场IPO定价模型及其实证研究[J].统计与信息论坛.2009,24(1):58-61
[18] Reilly F.and Hatfield K..Investor experience with new stock issue[J].Financial Analysts Journal 1969,25:73-80
[19] Ibbotson R..Price performance of common stock new issues[J].Journal of Financial Economics 1975,2:235-272
[20] Rock K..Why new issues are underpriced[J].Journal of Financial Economics,1986,15:187-212
[21] Beatty P..and Ritter R..Investment banking,reputation,and the underpricing of initial public offerings[J].Journal of Financial Economics 1986,15,213-232
[22] Baron P.and Holmstrom B..The investment banking contract for new issues under a symmetric information: delegation and the incentive problem[J].Journal of Finance,1980,35,1115-1138
[23] Biais B.,Bossaerts P.and Rochet J..An optimal IPO mechanism[J].Review of Economic Studies,2002,69,117-146.
[24] Vapnik V.,Golowich S. and Smola A..Support vector machine for function approximation,regression estimation,and signal processing[J].Advances in Neural Information Processing Systems,1996,9:281-287
[25]张铃.基于核函数的SVM机与三层前向神经网络的关系[J].计算机学报.2002,25(7):696-700
[26] Cortes C. and Vapnik..Support vector networks[J].Machine Learning,1995,20: 273-297
[27] Osuna E.,Freund R. and Girosi F..An improved training algorithm for support vector machines[A].Neural Networks for Signal ProceedingⅦ-Proceedings of the 1997 IEEE Workshop[C],IEEE,1997:267-285
[28] Platt C..Fast training of support vector machines using sequential minimal optimization[A].Schlkoph B..Advances in Kernel Method-Support Vector Learning[C].Cambridge,MIT Press,1998:185-208
[29] Joachims T. . Making large-scale SVM learning practical[A] . Schlkoph B..Advances in Kernel Method-Support Vector Learning[C].Cambridge,MIT Press,1999:169-184
[30] Suykens K. and Vandewale J. . Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300
[31] Chih-Chung Chang and Chih-Jen Lin.LIBSVM: a library for support vector machines,http://www.csie.ntu.edu.tw/~cjlin/libsvm,2001
[32] Chih-Wei Hsu,Chih-Chung Chang and Chih-Jen Lin.A Practical Guide to Support Vector Classification.http://www.csie.ntu.edu.tw/~cjlin,2003
[33] Mukherjee S.,Osuna E. and Girosi F.Nonlinear prediction of chaotic time series using support vector machine[A].Institute of Electrical and Electronics Engineers.In Proceedings of the IEEE Workshop on Neural Networks for Signal Processing[C].New York Press,1997:511–519
[34] Tay F. and Li-juan Cao.Application of support vector machines in financial time series forecasting[J].Omega the International Journal of Management Science,2001,29:309-317
[35] Tay F. and Li-juan Cao.Modified support vector machines in financial time series forecasting[J].Neurocomputing,2002,48:847-861
[36] Kim Kyoung-jae . Financial time series forecasting using support vector machines[J].Neurocompution,2003,55:307-319
[37] Ping-feng Pai and Chih-sheng Lin.A hybrid ARIMA and support vector machines model in stock price forecasting[J].Omega the International Journal of Management Science,2005(6):497-505
[38] Ping-Feng Pai and Wan-Ru Wei.Predicting movement directions of stock index futures by support vector models with data preprocessing[A] . Industrial Engineering and Engineering Management 2007 IEEE International Conference[C].2007:169-173
[39] Ruud S.Underwriter price support and the IPO under pricing puzzle[J].Journal of Financial Economics,1993,34:135-151
[40] Wun-Hua Chen and Jen-Ying Shih.Comparison of support-vector machines and back propagation neural networks in forecasting the six major Asian stock markets[J].International Journal of Electronic Finance,2006(1):49-67
[41] Kuan-Yu Chen.Evolutionary support vector regression modeling for TaiwanStock Exchange Market Weighted Index forecasting[J].Journal of American Academy of Business,2006(8):241-247
[42] Tinic M.Anatomy of Initial public offerings of common stock[J].Journal of Finance,1988,43:789-822
[43] Sheng-Hsun Hsu,Po-An Hsieh and Ting-Chih Chih.A two-stage architecture for stock price forecasting by integrating self-organizing map and support vector regression[J].Expert Systems with Applications,2009,36(4):7947-7951
[44]孙延风,梁艳春.支持向量机的数据依赖型核函数改进算法[J].吉林大学学报,2003,41(7):329-333
[45]李立辉.基于SVR的金融时间序列预测[J].计算机工程与应用,2005,30:221-224
[46]王彦峰,高风.基于支持向量机的股市预测[J].计算机仿真,2006,23(11):256-258
[47]张玉川,张作泉.支持向量机在股票价格预测中的应用[J].北京交通大学学报,2007,31(6):73-76
[48]隋学深,齐中英.基于多尺度特征和支持向量机的股市趋势预测[J].哈尔滨工业大学学报,2008,10(4):256-258
[49]李蓬宁.基于支持向量机的股市组合预测模型研究[J].经济研究导刊,2009,(22): 38-42
[50]陈圣滔,孙玉秋.统计模型在证券投资分析中的应用[J].技术经济与管理研究,2005,(2):37-38
[51] Sharp W.Capital Asset Prices:A Theory of Market Equilibrium under Conditions of Risks[J].Journal of Finance,1964,19:425-442
[52] Markowitz M.Portfolio Selection[J].Journal of Finance,1952,7:77-91
[53] Chi-Jie Lu,Tian-Shyug Lee and Chiu Chih-Chou.Financial time series forecasting using independent component analysis and support vector regression[J].Decision Support Systems,2009,47(2):115-125
[54]田翔,邓飞其.精确在线支持向量回归在股指预测中的应用[J].计算机工程,2005,31(22):18-20
[55]陈诗一.汇率预测:一个新的非参数支持向量回归方法[J].数量经济技术经济研究,2007(5):143-151
[56] Brennan J.and Franks J..Underpricing,ownership and control in initial public offerings of equity securities in the U.K. [J].Journal of Financial Economics,1997,45:391-413
[57] Mello S.and Parsons E..Going public and the ownership structure of the firm[J].Journal of Financial Economics,1998,49:79-109
[58] Ljungqvist A.,Nanda V.and Singh R..Hot markets,investor sentiment,and IPO pricing[J].Journal of Business,2006,79:166-1702
[59] Dorn D..Does sentiment drive the retail demand for IPOs?[J].Journal of Finance,,2003,12:22-26.
[60] Oehler A.,Rummer M.and Smith P.N..IPO pricing and the relative importance of investor sentiment-evidence from Germany[J].2004,12:15-18.
[61]张继强,周勇,张秉麟.IPO折价、逆向选择与分离均衡假说[J].预测,2003,4:46-61.
[62]周孝华,胡国生,苟思.中国股市IPOs高抑价的噪声分析[J].软科学,2005,19:30-33.
[63]王春峰,赵威,房振明.新股流动性风险及其与IPO抑价的关系[J].统计与决策,2006,24:21-25.
[64]夏国恩,邵培基.改进SVR在金融时间序列预测中的应用[J].金融理论与实践,2008,(11):95-98
[65]夏国恩,曾绍华,金炜东.支持向量回归机在铁路客运量时间序列预测中的应用[J].计算机应用研究,2006,(10):186-188
[66]夏国恩,金炜东,张葛祥.改进SVR及其在铁路客运量预测中的应用[J].西南交通大学学报,2007,42(4):494-498
[67]冯兴杰,魏新,王亚楼.基于支持向量回归的旅客吞吐量预测研究[J].计算机工程,2005,31(14):172-173
[68]李博,吴世农.中国股市新股发行(IPOS)的初始收益率研究[J].南开管理评论,2000,5,31-36
[69]杜莘,梁洪昀,宋逢明.中国A股市场首日收益率研究[J].管理科学学报,2001,4:55-61
[70]王春峰,姚锦.新股价值低估的随机前沿分析[J].系统工程,2002,22:22-23
[71]刘力,王汀汀.不应忽略股票的流通权价值——兼论中国股票市场的二元股权结构问题[J].管理世界,2003,9:46-51
[72]朱南,卓贤.对我国股票首次公开发行抑价的实证研究[J].财经科学,2004,2:32-36
[73]李翔,阴永晟.发行管制变迁下的中国股市IPO首日回报率研究[J].经济科学,2004,3:43-53
[74]刘煌辉,熊鹏.股权分置、政府管制和中国IPO抑价[J].经济研究,2005,5:85-95
[75]曹凤岐,董秀良.我国IPO定价合理性的实证分析[J].财经研究,2006,32:1-3
[76]张锦莉.沪深股市IPO溢价现象的研究[D].暨南大学硕士学位论文,2008
[77]刘路.中国IPO首日超额收益理论与实证分析[D].复旦大学硕士学位论文,2008
[78]唐勇.发售机制、信息偏倚与IPO抑价[D].浙江大学硕士学位论文,2009
[79]李允默.我国新股上市首日溢价研究[D].武汉科技大学硕士学位论文,2009
[80]李冬琴,王丽铮,王呈方.一种新的支持向量回归算法及其在集装箱吞吐量预测中的应用[J].水运工程,2007,(5):16-19
[81]董辉,傅鹤林,冷伍明.支持向量回归的时间序列回归与预测[J].系统仿真学报,2006,18(7):1785-1788
[82]邵信光,杨慧中,石晨曦.ε-不敏感支持向量回归在化工数据建模中的应用[J].东南大学学报(自然科学版),2004,34:215-218
[83]黎展求,朱栋华,刘冬岩.基于支持向量回归和小波包的供热负荷预测[J].暖通空调,2007,37(2):1-5
[84]徐红敏,刘亚禄.改进的支持向量回归机及其在水质预测中的应用[J].统计与决策,2006,(23):12-13
[85]魏宏业,吕永波,何崴.基与数据挖掘的企业合作伙伴的寻求方法研究.管理科学学报,2004,7(1):60-64
[86] Vapnik V..统计学习理论的本质[M].北京:清华大学出版社,2000:11-155