基于BP神经网络的股价预测模型应用分析

发布时间：2018-04-12 09:39

本文选题：股价预测 + BP神经网络　；参考：《暨南大学》2017年硕士论文

【摘要】：作为金融业不可或缺的一部分,股票投资在人们的生活中呈现愈来愈重要的态势。股价的波动直接影响着股市的稳定以及金融市场与国民经济的健康发展。成功的股价和趋势预测有助于投资者获利,也有助于政府部门提供及时合理的市场指导与监管。因此,如何预测股票市场是金融界存在的一个重要而有价值的课题。神经网络模拟人体神经元功能,可以从股市中提取历史数据和相关信息来推断股价未来发展趋势,具有较强的自组织与自适应的能力。因此,神经网络适合于处理具有非线性时间序列特征的股价预测问题。本文提出了一种基于BP神经网络针对股价预测问题的综合模型。从影响股价的市场面因素中讨论并确定了21个针对股价预测的评价指标,通过主成分分析方法降维至6个输入变量;同时通过思维进化算法优化网络初始连接权值和阈值;探讨了网络的拓扑结构、隐含层节点个数、学习率、激活函数和训练函数、样本数据的选取和预处理等问题;最后建立对比模型并通过多次运行,得到综合模型拥有最高的预测精度和稳定性。理论分析和实验结果表明,本文建立的基于BP神经网络的股价预测综合模型是可行且有效的,有着良好的应用前景。
[Abstract]:As an indispensable part of the financial industry, stock investment is becoming more and more important in people's lives.The fluctuation of stock price directly affects the stability of stock market and the healthy development of financial market and national economy.Successful stock price and trend forecasts help investors make profits and help government departments to provide timely and reasonable market guidance and regulation.Therefore, how to predict the stock market is an important and valuable issue in financial circles.Neural network simulates human neuron function and can extract historical data and relevant information from stock market to infer the future development trend of stock price. It has strong ability of self-organization and self-adaptation.Therefore, the neural network is suitable to deal with the stock price forecasting problem with nonlinear time series.This paper presents a comprehensive model for stock price forecasting based on BP neural network.This paper discusses and determines 21 evaluation indexes of stock price prediction from the factors affecting stock price, reduces the dimension to 6 input variables by principal component analysis, and optimizes the initial connection weight and threshold of the network through the thinking evolution algorithm.The topological structure of the network, the number of hidden layer nodes, the learning rate, the activation function and the training function, the selection and preprocessing of the sample data and so on are discussed.The synthetic model has the highest prediction accuracy and stability.The theoretical analysis and experimental results show that the synthetic model of stock price prediction based on BP neural network is feasible and effective and has a good application prospect.
【学位授予单位】：暨南大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP183;F832.51

【参考文献】