基于BP神经网络的股票价格预测  

Stock price prediction based on BP neural network

作　　者：曾丽芳 李丽萍 江绍萍[1] Zeng Li-fang;Li Li-ping;Jiang Shao-ping(School of Mathematics and Computer Science,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）》2023年第5期658-665,共8页Journal of Yunnan Minzu University:Natural Sciences Edition

摘　　要：影响股票价格变动的因素有很多,且股票数据具有高度的非线性和时变性等特征,因而采用经典线性时间序列模型可能无法完全提取非线性部分的信息.针对这一问题,建立了BP神经网络模型、PCA-BP神经网络模型、GA-BP神经网络模型和ARIMA(6,1,6)模型对上证综合指数的收盘价格进行预测.计算各预测模型下的统计指标RMSE和MAE,并对4个模型进行对比分析.结果表明,GA-BP神经网络预测模型与其它三种模型相比具有更小的误差,也就是说GA-BP神经网络预测模型对上证综合指数的收盘价格预测效果更好.Due to there are many factors affecting the change of stock price,and the stock price data has the characteristics of high nonlinearity and time variability.Therefore,the classical linear time series model may not be able to completely extract the information of the nonlinear part.To solve this problem,BP neural network model,PCA-BP Neural Network Model,GA-BP Neural Network Model and ARIMA(6,1,6)model are established to predict the closing price of Shanghai Composite Index.Calculate the statistical indexes MRSE and MAE under each prediction model,and compare and analyze the four models.The results show that the GA-BP neural network prediction model has less error than the other three models,that is,the GA-BP neural network prediction model has better prediction effect on the closing price of Shanghai Composite Index.

关 键 词：上证综合指数 BP神经网络 ARIMA模型 主成分分析 遗传算法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]