一种新的BP神经网络预测金融相关系数
    点此下载全文
引用本文:杨兴华?覮,吴伟,王林浩.一种新的BP神经网络预测金融相关系数[J].计算技术与自动化,2019,(1):20-24
摘要点击次数: 680
全文下载次数: 0
作者单位
杨兴华?覮,吴伟,王林浩 (中国卫星海上测控部江苏 无锡 214434) 
中文摘要:提出了一种改进的神经网络预测两个金融时间序列的交叉相关(cross-correlation)系数。为了得到金融数据集的波动,对传统BP神经网络进行了改进得到了一种指数BP神经网络,通过计算输入向量与其权值向量之间的点积,不仅对每个神经单元进行局部信息处理,还通过在输入向量的指数型函数及其相应的新权向量之间增加点积来进行处理。该预测模型改进了神经网络的激活函数,并对特定输入输出变量的交叉相关预测进行了探讨。实验证明,所提模型有利于提高预测精度。
中文关键词:人工智能  神经网络  预测  波动率  神经元
 
A Novel BP Neural Network Forecasting for Financial Cross-correlation Relationship
Abstract:An improved neural network is proposed to predict the cross-correlation coefficient of two financial time series. In order to obtain the volatility of the financial data set,an exponential BP neural network is obtained for the improvement of the traditional BP neural network,which information is not only processed locally in each neural unit by computing the dot product between its input vector and its weight vector,but also processed by adding the dot product between its exponential type function of the input vector and its corresponding new weight vector. The prediction model improves the activation function of the neural network and discusses the cross correlation prediction of the specific input and output variables. The result shows that the model is advantageous in increasing the predicting precision.
keywords:artificial intelligence  neural network  predict  volatility  neuron
查看全文   查看/发表评论   下载pdf阅读器