基于投资者情绪和栈式自编码器的股价预测模型  

Stock price prediction model based on investor sentiment and stacked denoising autoencoders

作　　者：蔡俊杰 王爱银[1] CAI Junjie;WANG Aiyin(School of Statistics and Data Science,Xinjiang University of Finance and Economics,Urumqi 830000,China)

机构地区：[1]新疆财经大学统计与数据科学学院,乌鲁木齐830000

出　　处：《哈尔滨商业大学学报(自然科学版)》2025年第1期120-128,共9页Journal of Harbin University of Commerce:Natural Sciences Edition

基　　金：国家社科基金一般项目《西部民族地区提升基本公共服务可及性和均衡性路径研究》(23BJL122)。

摘　　要：为提高股价预测的准确性,通过非线性组合的方法,构造了一种融合投资者情绪和栈式去噪自编码器(SDAE)和LSTM组合模型.通过情感分析(SA)提取的情感指数和SDAE提取的股票高质量特征被用作LSTM模型的输入.基于Python开发环境对恒生指数(HSI)进行了研究,实验结果表明,所提方法的预测性能优于其他对比方法,其平均绝对误差(MAPE)、R^(2)和方向准确度(DA)值分别达到1.12%、0.92和84.93%,具有准确度较高的预测能力.To enhance the accuracy of stock price forecasting,a combined model integrating investor sentiment,Stacked Denoising Autoencoder(SDAE),and LSTM was constructed using a nonlinear combination approach.The sentiment index extracted by the sentiment analysis(SA)and the high-quality stock features extracted by SDAE were utilized as inputs for the LSTM model.Within the Python development environment,an empirical study on the Hang Seng Index(HSI)was conducted.The experimental results demonstrated that the proposed method outperformed other comparative methods.The mean absolute percentage error(MAPE),R^(2),and directional accuracy(DA)were found to be 1.12%,0.92,and 84.93%,respectively,indicating a high predictive accuracy of the proposed model.

关 键 词：股价预测 投资者情绪 栈式去噪自编码器 长短期记忆网络 非线性组合 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]