基于主成分分析和长短期记忆网络的股票价格预测

Stock Price Prediction Based on Principal Component Analysis and Long Short-Term Memory Network

摘要: 运用神经网络技术，建立基于主成分分析的长短期记忆神经网络(PCA-LSTM)模型并对股票开盘价格进行预测。实验采用五粮液(000858)股票，首先，利用主成分法对该股票的多个指标进行特征提取，然后利用提取的主成分建立LSTM神经网络模型，并与PCA-Elman、LSTM模型对比，结果发现PCA-LSTM模型的预测结果更好一些。

Abstract: Using the neural network technology, the long and short term memory neural network (PA-LSTM) model based on principal component analysis was established and the stock opening price was predicted. In the experiment, wuliangye (000858) stock was used. First, multiple indexes of the stock were extracted by principal component method. Then, LSTM neural network model was established by using the extracted principal component and compared with THE PCA-Elman and LSTM models. The results show that the prediction results of PCA-LSTM model are better.