基于信度隐马尔可夫过程的股票指数趋势预测方法  

The Stock Index Trend Prediction Method Based on Belief Hidden Markov Process

作　　者：王璐璐 刘邱云[3] 赖岳 WANG Lulu;LIU Qiuyun;LAI Yue(School of Finance,Jiangxi University of Finance and Economics,Nanchang Jiangxi 330013,China;School of Economics and Management,Jiangxi Flight University,Nanchang Jiangxi 330088,China;School of Mathematics and Statistics,Jiangxi Normal University,Nanchang Jiangxi 330022,China)

机构地区：[1]江西财经大学金融学院,江西南昌330013 [2]江西飞行学院经济与管理学院,江西南昌330088 [3]江西师范大学数学与统计学院,江西南昌330022

出　　处：《江西师范大学学报(自然科学版)》2024年第6期597-603,共7页Journal of Jiangxi Normal University(Natural Science Edition)

基　　金：江西省文化艺术科学规划课题(YG2022175)资助项目.

摘　　要：由于股票指数序列具有高频性、非线性和长记忆性等特点,且投资者行为的不确定性,所以其预测工作甚为艰难.该文结合证据理论和Pignistic概率转换方法,研究了一种新的马尔可夫模型转移矩阵计算方法,并应用于股票指数趋势预测,提出了基于信度隐马尔可夫过程的股票指数趋势预测方法,实证结果表明该方法的预测效果较好.Due to the high-frequency,non-linear,and long-term memory characteristics of stock index sequences,as well as the uncertainty of investor behavior,their prediction work is very difficult.The evidence theory and logistic probability transformation methods are combined to provide a new method for calculating Markov model transition matrix.Based on this,the stock index trend prediction method based on belief hidden Markov process is proposed,and the experimental results show that the prediction effect is good.

关 键 词：隐马尔可夫模型 证据理论 信度函数 Pignistic概率 股票指数预测 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]