【摘要】 基于LSTM神经网络模型进行股票价格的预测研究。利用开市以来的七千多条上证综合指数数据,使用长短期记忆(LSTM)神经网络模型对上证综合指数进行预测分析,并将其预测结果与使用BP神经网络模型、CNN、RNN、GRU网络模型的预测结果进行对比。结果显示LSTM神经网络模型的预测效果最好,其评价指标中的平均绝对误差(MAE)为0.015 799,均方误差(MSE)为0.000 450,平均绝对百分比误差(MAPE)为0.019 867,预测误差低于其他模型;其预测值和真实值之间的相关系数为0.995 7,表明预测值和真实值的拟合程度较高。更多还原

【Abstract】 LSTM neural network model is used to predict stock prices. Based on more than 7 000 Shanghai Composite Index data since the opening of the market,long short-term memory( LSTM) neural network model is used to conduct the prediction analysis,and the prediction results are compared with those of the BP neural network model,CNN model,RNN model and GRU network model. The results show that the LSTM neural network model has the best prediction effect. The average absolute error( MAE) of the evaluation index is 0. 015 799,the mean square error( MSE) is 0. 000 450,the average absolute percentage error( MAPE) is 0. 019 867,and the prediction error is lower than other models. The correlation coefficient between the predicted value and the true value is 0. 995 7,indicating that the predicted value and true value have a high degree of fitting.更多还原