摘要
运用递归神经网络,并结合主成分分析方法建立基于主成分分析的递归神经网络(PCA-RNN)预测模型.实验采用玉米股票价格指数,首先,利用主成分法对玉米指数的多个指标进行特征提取,然后利用提取的主成分建立3种神经网络模型,并对开盘价进行预测,最后与ARIMA模型进行比较分析.结果表明PCA-RNN模型取得了较好的效果,更加适用于股票价格的短期预测,可以为决策者提供一定的参考.
Using recurrent neural network and principal component analysis method to establish principal component analysis and recurrent neural network(PCA-RNN) prediction model.The experiment adopts the corn stock price index.Firstly,the principal component method is used to extract the features of the corn index.Then,three neural network models are established by using the extracted principal components,and the opening price is predicted.Finally,the results are compared and analyzed with those from the ARIMA model.The results show that the PCA-RNN model has achieved good results and is more suitable for short-term forecasting of stock prices,which can provide some reference for decision makers.
作者
孙德山
任靓
SUN Deshan;REN Liang(School of Mathematics,Liaoning Normal University,Dalian 116029,China)
出处
《辽宁师范大学学报（自然科学版）》
CAS
2019年第3期301-306,共6页
Journal of Liaoning Normal University:Natural Science Edition
基金
辽宁省自然科学基金资助项目(2019-ZD-0471)
