您的位置:首页  > 论文页面

对角型广义RBF神经网络在股票价格预测中的应用

发表时间:2014-07-15  浏览量:1587  下载量:729
全部作者: 刘双龙,马尽文
作者单位: 北京大学数学科学学院,数学及其应用教育部重点实验室
摘 要: 将对角型广义径向基函数(radial basis function,RBF)神经网络应用于股票价格时间序列的分析与预测。在网络设计上,采用基于偏峰度准则的动态分合期望最大化(expectation maximization,EM)算法进行网络隐单元个数的确定,并同时完成参数初始值的选取;在网络学习上,采用同步最小均方误差(least mean square,LMS)学习算法进行参数学习和确定。实验结果表明:对角型广义RBF神经网络能够有效地进行股票价格预测,具有较好的应用前景。
关 键 词: 应用数学;径向基函数神经网络;期望最大化算法;股票价格;时间序列预测
Title: Application of diagonal generalized RBF neural network to stock price prediction
Author: LIU Shuanglong, MA Jinwen
Organization: Key Lab of Mathematics and Applied Mathematics, Ministry of Education, School of Mathematical Sciences, Peking University
Abstract: In this paper, the diagonal generalized radial basis function (RBF) neural network is applied to analyzeing and predicting the time series of stock prices. Specifically, the dynamic split-and-merge expectation maximization (EM) algorithm with the kurtosis and skewness criterion is utilized to determine the number of hidden units as well as the initial values of the parameters for the network design, while the synchronous least mean square error (LMS) learning algorithm is implemented for the final learning of the parameters in the network. It is demonstrated by the experiment results that the diagonal generalized RBF neural network can be successfully applied to the prediction of stock prices, which shows a good application prospect on this aspect.
Key words: applied mathematics; radial basis function neural network; expectation maximization algorithm; stock price; time series prediction
发表期数: 2014年7月第13期
引用格式: 刘双龙,马尽文. 对角型广义RBF神经网络在股票价格预测中的应用[J]. 中国科技论文在线精品论文,2014,7(13):1296-1306.
 
2 评论数 0
暂无评论
友情链接