基于模糊信息粒化和支持向量机的股票价格回归预测  

Stock price regression prediction based on fuzzy information granulation and a support vector machine

作　　者：郑明[1] 李娌芝 官心果 杨柱元[1] ZHENG Ming;Li Li-zhi;GUAN Xin-guo(School of Mathematics and Computer Sciences,Yunnan Minzu University,Kunming 650500,China)

机构地区：[1]云南民族大学数学与计算机科学学院,云南昆明650500

出　　处：《云南民族大学学报（自然科学版）》2018年第6期517-524,共8页Journal of Yunnan Minzu University:Natural Sciences Edition

基　　金：国家自然科学基金(11361076)

摘　　要：对股票进行投资时,由于股票数据具有较大的不稳定性,往往大多数时候无法对其进行精确的预测,而对其变化趋势和变化空间进行预测尤为重要,当利用支持向量机对股票数据进行回归预测时,只能得到具体点的预测值,不能预测股票数据短期的变化趋势,因此本文将模糊信息粒化和支持向量机相结合,对股票数据未来5天的变化趋势进行了预测研究,实验表明该方法具有理想的效果.When investing in stocks, investors want to predict the trading results in the next few days as well as their trend and changes. The traditional time-series regression method is used to predict the stock data, but it only gives the concrete prediction value of the stock data, and often fails to predict the changing trend and scope. Therefore, in this paper, a support vector machine based on fuzzy information granulation is given, then the stock price is regressed, and the same stock price is predicted by a single support vector machine. Finally, the forecasting analysis of the stock price is carried out before the integration of the results of the two for a correct prediction. The paper takes the stock data of the previous years as a sample for simulation training, uses the cross-validation method to optimize the parameters of SVM. First, a single support vector regression machine is used to predict the stock price in the up-coming five trading days; at the same time, the support vector machine based on fuzzy information granulation is used to predict the changing trend of the stock price in the next five trading days,and finally, the consistency of their prediction results is compared. The feasibility and effectiveness of the algorithm are verified by prediction experiments on stocks, and, combined with the traditional SVM regression method, the prediction accuracy of the stock is more accurate.

关 键 词：支持向量机 信息粒化 回归预测 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] F832.5[自动化与计算机技术—控制科学与工程]