基于IndRNN-LSTM模型的股票价格预测  

Stock Price Prediction Based on Independently Recurrent Neural Network and Long Short-Term Memory Network

作　　者：扈文 

机构地区：[1]辽宁师范大学数学学院,辽宁 大连

出　　处：《应用数学进展》2022年第1期209-218,共10页Advances in Applied Mathematics

摘　　要：结合独立循环神经网络和长短期记忆网络建立IndRNN-LSTM模型,选取道琼斯指数的18个指标,对道琼斯指数的开盘价格进行预测。另外,分别采用CNN-LSTM、IndRNN、LSTM、SVM、BP和CNN神经网络模型对开盘价格进行预测,并将七种模型的预测结果进行比较,结果表明IndRNN-LSTM模型的预测精度较高,能更好地预测股票的走势。IndRNN-LSTM model is established by combining independently recurrent neural network and long short-term memory network. This paper selects 18 indexes of Dow Jones index to predict the opening price of Dow Jones index. CNN-LSTM, IndRNN, SVM and LSTM neural network models are also used to predict the opening price, and the prediction results of the five models are compared. The results show that the IndRNN-LSTM model has high prediction accuracy and can better predict the stock trend.

关 键 词：独立循环神经网络 长短期记忆网络 道琼斯指数 

分 类 号：F832.5[经济管理—金融学]