BP神经网络的改进及其在股票预测中的应用  

The Improvements of BP Neural Network and Its Application in Stock Prediction

作　　者：王亿楷[1] 赖国明[1] 

机构地区：[1]韩山师范学院数学与信息技术系,广东潮州521041

出　　处：《韩山师范学院学报》2008年第6期39-43,共5页Journal of Hanshan Normal University

摘　　要：研究了BP神经网络,对BP算法的权值初始化进行改进,使得其收敛速度更快,并建立了基于BP神经网络的股价预测模型.最后以深发展A的收盘价为例,分析改进后的BP神经网络模型的预测效果.This article studies the BP neural network, makes the improvement to the BP algorithm＇s initial weight, causes its convergence rate to be quicker, and has established stock price forecast model based on the BP neural network. Finally, taking Shenzhen Development A＇s closing price as example data; it analyzes the forecasting effect of the improved BP neural network model.

关 键 词：神经网络 BP算法 股票预测 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]