基于Cart决策树与boosting方法的股票预测  

A Stock Prediction Model Based on Cart and Boosting Algorithm

作　　者：王禹 陈德运[1] 唐远新[1] WANG Yu;CHEN De-yun;TANG Yuan-xin(School of Computer Science and Technology,Harbin University of Science and Technology,Harbin 150080,China)

机构地区：[1]哈尔滨理工大学计算机科学与技术学院,哈尔滨150080

出　　处：《哈尔滨理工大学学报》2019年第6期98-103,共6页Journal of Harbin University of Science and Technology

基　　金：国家自然科学基金(69572153,60972127).

摘　　要：针对股票预测模型的准确度不够高,存在过拟合及欠拟合等问题,在现有股票预测方法分析的基础上,给出了一种基于Cart决策树与boosting方法的股票预测方法。该方法针对现有预测模型在数据纵向性方面考虑较少,额外增添近10日均价及转手率两个纵向变化指标以提高股票走势预测的准确性;并且以Cart决策树方法为基础,采用boosting级联多棵决策树的方法建立股票模型来解决拟合度问题。预测实验结果表明,该方法对仪器仪表领域的股票预测效果较好,均方误差有所下降。The accuracy of the stock prediction model is notaccurate enough,for there are problems such as over-fitting and under-fitting.Based on the analysis of the existing stock prediction methods,this paper proposes a stock prediction method based on Cart decision tree and boosting.Current forecasting models have less consideration in data verticality,so we adds the average price in the past ten days and turnover rate to improve the accuracy of the stock prediction.The method is based on the Cart decision tree method.We use boosting to concatenation multiple decision tree to build a stock model to solve the over-fitting problem.The prediction experiment results show that this stock prediction has a good effect in the field of instruments and meters,and the mean-square error has been decreased.

关 键 词：股票预测 Cart树 BOOSTING算法 

分 类 号：TP81[自动化与计算机技术—检测技术与自动化装置]