基于Elman神经网络的时间序列股票价格短期预测  

Short-term Prediction on Time Series of Stock Price based on Elman Neural Network

作　　者：张慧斌[1] 高秀萍[2] 

机构地区：[1]忻州师范学院计算机科学与技术系,山西忻州034000 [2]中国人民财产保险股份有限公司德州分公司,山东德州253020

出　　处：《山西大同大学学报（自然科学版）》2011年第2期5-7,60,共4页Journal of Shanxi Datong University(Natural Science Edition)

摘　　要：Elman神经网络是一种典型的回归神经网络,比BP神经网络具有更强的计算和适应时变特性的能力,因而非常适用于预测股市这一类极其复杂的非线性动力学系统。文章给出一种基于Elman神经网络的股票市场建模、预测及决策方法,对浦发银行股价在时间序列上作了连续若干天的短期预测,实验结果取得较高的预测精度、较为稳定的预测效果和较快的收敛速度。这表明该预测模型对于个股价格的短期预测是可行和有效的。Elman neural network is a classical kind of recurrent neural network.It has greater ability to calculate than BP neural network.Thanks to its characteristic of adapting time variability,Elman neural network is very suitable to predict complicated nonlinear dynamics system such as stock market.This paper provides a stock market model and an approach to predict and decide based on Elman neural network.Meanwhile the author makes a short-term prediction on time series of PuFa Bank＇s stock price.The results of the experiment get higher precision,steadier forecasting effect and more rapid convergence speed.It shows that this kind of model is feasible and efficient to predict short-term stock price.

关 键 词：ELMAN神经网络 时间序列 股票价格预测 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]