基于CART和相似股的股票价格走势预测算法研究
Forecasting Algorithm Research of Stock Price Trend Based on CART and Similar Stock
DOI: 10.12677/CSA.2017.76071, PDF, HTML, XML, 下载: 1,913  浏览: 6,150 
作者: 刘建中：天津市国资委，天津；殷其威：南开大学计算机与控制工程学院，天津
关键词: 股票数据；滑动窗口；相似股票；CART；ARMA；Stock Data； Slide Window； Similar Stock； CART； ARMA
摘要: 目前预测股票价格大多都基于单支股票的历史价格数据，并试图找出其股价变化规律，训练出可预测价格的模型。但实际股票价格的波动会受众多社会实时因素和投资者行为的影响，因此基于历史数据的股票价格预测模型往往失效。为此，本文研究2100支股票中具有相似历史价格变化的股票，并基于时间序列窗口滑动获得用于预测模型的训练的数据和预测数据，使用决策树CART(Classification and Regression Trees)对预测结果数据做出判别。并与经典的时间序列分析模型ARMA(Auto regressive Moving Average Model)对股票价格走势预测的结果进行分析比较，实验结果验证了本文提出的方法预测结果的有效性。
Abstract: Most of the current forecast stock prices are based on a single stock historical price data, and try to find out the law of its stock price changes, training a model which can be used for price fore-casting. But the fluctuation of the stock price will be affected by many social real-time factors and investors’ behavior, so the stock price forecast model based on historical data often fails. For this reason, in this paper, we study stocks with similar historical changes in 2100 stocks, and obtain training data and forecast data for forecasting model based on time series window sliding, then use the CART (Classification and Regression Trees) to determine the predicted data. And compared with the result of ARMA (Auto Regressive Moving Average Model), the validity of the method is verified.
文章引用：刘建中, 殷其威. 基于CART和相似股的股票价格走势预测算法研究[J]. 计算机科学与应用, 2017, 7(6): 603-614. https://doi.org/10.12677/CSA.2017.76071

参考文献

[1] 杜建卫, 王超峰. 小波分析方法在金融股票数据预测中的应用[J]. 数学的实践与认识, 2008, 38(7): 68-75.
[2] 邓凯旭, 宋宝瑞. 小波变换在金融数据分析中的应用[J]. 数理统计与管理, 2006, 25(2): 215-219.
[3] Das G., Lin, D., Mannila, H., Renganathan, G. and Smyth, P. (1998) Rule Discovery from Time Series. Fourth Annual Conference on Knowledge Discovery and Data Mining (KDD), New York, 27-31August, 23-29.
[4] 李斌. 金融时间序列挖掘关键算法研究[D]: [博士学位论文]. 合肥: 中国科学技术大学, 2002.
[5] 辛治运, 顾明. 基于最小二乘支持向量机的复杂金融时间序列预测[J]. 清华大学学报:自然科学版, 2008, 48(7): 1147-1149.
[6] Cao, L. and Tay, F.E.H. (2001) Financial Forecasting Using Support Vector Machines. Neural Computing and Applications, 10, 184-192.
https://doi.org/10.1007/s005210170010
[7] Kimoto, T., Asakawa, K., Yoda, M. and Takeoka, M. (1990) Stock Market Prediction System with Modular Neural Networks. International Joint Conference on Neural, 1, 1-6.
https://doi.org/10.1109/ijcnn.1990.137535
[8] 易东云, 张维明. 金融数据挖掘中的非线性相关跟踪技术Nonlinear Correlation Tracking Technique in Data Mining of Financial Markets [J]. 软件学报, 2000, 11(12): 1581-1586.
[9] 汤胤. 时间序列相似性分析方法研究[J]. 计算机工程与应用, 2006, 42(1): 68-71.
[10] Breiman, L.I., Friedman, J.H., Olshen, R.A. and Stone .C.J. (1984) Classification and Regression Trees (CART). Biometrics, 40, 358.
[11] 王丽娜, 肖冬荣. 基于ARMA模型的经济非平稳时间序列的预测分析[J]. 武汉理工大学学报交通科学与工程版, 2004, 28(1): 133-136.
[12] 张海勇, 马孝江, 盖强. 非平稳信号的一种ARMA模型分析方法[J]. 电子与信息学报, 2002, 24(7): 992-996.
[13] 吴朝阳. 改进的灰色模型与ARMA模型的股指预测[J]. 智能系统学报, 2010, 5(3): 277-281.