基于高斯过程回归的上市股价预测模型
|
摘要
在新股上市价格的科学优化预测问题的研究中,由于金融数据复杂,特别是新股价格存在极强的无序性。传统股票价格预测方法只能采用线性变化规律进行准确预测,无法对非线性股票价格进行有效建模,降低股价预测精度。为了提高股票价格预测精度,提出一种高斯过程回归的新股上市价格预测模型,通过提取影响新股上市价格形成的指标因素,用其训练纳斯达克(NASDAQ)新股上市价格的历史数据,以粒子群算法优化高斯过程的超参数来预测新股上市价格。将8家公司的上市股票作为实例进行分析,预测结果表明,高斯过程回归的方法提高股票价格预测精度,能够有效地适用于新股上市价格预测。
Scientific price prediction for initial public offerings(IPO) has occupied a very important position in the field of financial investment.The traditional stock price prediction methods can only accurately predict stock prices for linear variation.They are unable to model nonlinear stock prices,which decreases the prediction accuracy of stock prices.In order to improve the prediction accuracy of stock prices,this paper proposed a price forecasting approach for IPO based on Gaussian process regression.It trains the history prices of NASDAQ IPO through extracting the index factors influencing the stock prices.It uses the hyper-parameters of Gaussian process that are optimized by particle swarm optimization(PSO) algorithm to predict IPO price.It analyzes the marketing stocks of eight companies as test samples.The predictive results show that the Gaussian process regression method improves the prediction accuracy of stock prices and is effectively applied to the prices for IPO.
Scientific price prediction for initial public offerings(IPO) has occupied a very important position in the field of financial investment.The traditional stock price prediction methods can only accurately predict stock prices for linear variation.They are unable to model nonlinear stock prices,which decreases the prediction accuracy of stock prices.In order to improve the prediction accuracy of stock prices,this paper proposed a price forecasting approach for IPO based on Gaussian process regression.It trains the history prices of NASDAQ IPO through extracting the index factors influencing the stock prices.It uses the hyper-parameters of Gaussian process that are optimized by particle swarm optimization(PSO) algorithm to predict IPO price.It analyzes the marketing stocks of eight companies as test samples.The predictive results show that the Gaussian process regression method improves the prediction accuracy of stock prices and is effectively applied to the prices for IPO.
引文
[1]孔玉生,王忠杰.新股上市首日定价预测模型研究[J].中国管理科学,2003,11(1):6-9.
[2]RSharda,R B Patil.A connectionist approach to time series pre-diction:an empirical test[J].Journal of Intelligent Manufactur-ing.1992,3(1):317-323.
[3]L Kryzanowski,M Galler,D W Wright.Using artificial neural net-works to pick stocks[J].Financial Analysts Journal.1993,49
[4]R R Trippi,E Turban.Neural networks in finance and investingusing artificial intelligence to improve real-world performance[M].Chicago:Probus Company,1993.
[5]杨新斌,黄晓娟.基于支持向量机的股票价格预测研究[J].计算机仿真,2010,27(9):302-305.
[6]B A Jain,B N Nag.Artificial neural network models for pricinginitial public offerings[J].Decision Sciences.1995,26(3):283-302.
[7]G Montagna.Pricing derivatives by path in integral and neural net-work[J].Physica A.2003,324(2):189-195.
[8]李桢,徐凌宇.基于阶段评价的BP及在股价预测中的应用[J].计算机仿真,2006,23(12):276-280.
[9]周世昊,林苍祥,倪衍森.基于遗传算法和神经网络的新股上市价格预测法[J].计算机工程,2007,33(22):9-11.
[10]S ABillings.Radial basis function network configuration using ge-netic algorithms[J].Neural Networks,1995,8(6):877-890.
[11]R SSexton,R E Dorsey,J D Jihnson.Toward global optimizationfor artificial neural networks:a comparison of the genetic algo-rithm and back propagation[J].Decision Support Systems,1998,22(2):171-185.
[12]夏禹.人工神经网络在首次公开募股价格预测中的运用[J].微型电脑应用,2010,26(6):51-54.
[13]王唯贤,陈利军.股票价格预测的建模与仿真研究[J].计算机仿真,2012,29(1):344-347.
[14]甘昕艳,张钰玲,潘家英.基于股价指数预测的仿真研究[J].计算机仿真,2010,27(10):297-300.
[15]M Seeger.Gaussian processes for machine learning[J].Interna-tional Journal of Neural System,2004,14(2):69-106.
[16]S BBelhouari,A Bermak.Gaussian process for nonstationarytimes series prediction[J].Computational Statistics&Data A-nalysis,2004,47(4):705-712.
[17]R Poli,J Kennedy,T Blackwell.Particle swarm optimization:anoverview[J].Swarm Intelligence,2007,1(1):33-57.
[2]RSharda,R B Patil.A connectionist approach to time series pre-diction:an empirical test[J].Journal of Intelligent Manufactur-ing.1992,3(1):317-323.
[3]L Kryzanowski,M Galler,D W Wright.Using artificial neural net-works to pick stocks[J].Financial Analysts Journal.1993,49
[4]R R Trippi,E Turban.Neural networks in finance and investingusing artificial intelligence to improve real-world performance[M].Chicago:Probus Company,1993.
[5]杨新斌,黄晓娟.基于支持向量机的股票价格预测研究[J].计算机仿真,2010,27(9):302-305.
[6]B A Jain,B N Nag.Artificial neural network models for pricinginitial public offerings[J].Decision Sciences.1995,26(3):283-302.
[7]G Montagna.Pricing derivatives by path in integral and neural net-work[J].Physica A.2003,324(2):189-195.
[8]李桢,徐凌宇.基于阶段评价的BP及在股价预测中的应用[J].计算机仿真,2006,23(12):276-280.
[9]周世昊,林苍祥,倪衍森.基于遗传算法和神经网络的新股上市价格预测法[J].计算机工程,2007,33(22):9-11.
[10]S ABillings.Radial basis function network configuration using ge-netic algorithms[J].Neural Networks,1995,8(6):877-890.
[11]R SSexton,R E Dorsey,J D Jihnson.Toward global optimizationfor artificial neural networks:a comparison of the genetic algo-rithm and back propagation[J].Decision Support Systems,1998,22(2):171-185.
[12]夏禹.人工神经网络在首次公开募股价格预测中的运用[J].微型电脑应用,2010,26(6):51-54.
[13]王唯贤,陈利军.股票价格预测的建模与仿真研究[J].计算机仿真,2012,29(1):344-347.
[14]甘昕艳,张钰玲,潘家英.基于股价指数预测的仿真研究[J].计算机仿真,2010,27(10):297-300.
[15]M Seeger.Gaussian processes for machine learning[J].Interna-tional Journal of Neural System,2004,14(2):69-106.
[16]S BBelhouari,A Bermak.Gaussian process for nonstationarytimes series prediction[J].Computational Statistics&Data A-nalysis,2004,47(4):705-712.
[17]R Poli,J Kennedy,T Blackwell.Particle swarm optimization:anoverview[J].Swarm Intelligence,2007,1(1):33-57.