融合媒体信息和信号分解的股票市场深度学习预测

doi:10.11896/jsjkx.230600102

摘要/Abstract

摘要： 对股票市场未来回报和风险的精确预测不仅能够帮助理性投资者更加合理有效地进行投资,也能够为政策制定者和投资者提供有用的指导。利用金融新闻标题文本,通过词嵌入模型和机器学习等文本分析方法,构建考虑新闻累积效应的投资者时闻累积情绪指数表征投资者情绪;以上证指数为例,采用变分模式分解(VMD)方法将指数波动数据分解为各种内在固有模式进行实证分析。最后,引入双向门控循环单元(BiGRU)作为深度学习模型进行股票预测。结果表明,投资者情绪指数显著影响上证指数波动,并且积极情绪和消极情绪的影响是不对称的;考量投资者情绪指标进行信号分解,能够有效提高股票的预测性能,相对于单纯分析股票时间序列的 BiGRU预测模型,VMD-BiGRU模型的MAE,RMSE,RMSPE,MAPE等指标降低超过30%;在基准场景下,VMD-BiGRU模型性能优于多个计量经济模型和机器学习模型,对于收益率和波动率预测的MAE,RMSE,RMSPE,MAPE等指标普遍降低超过40%;模型在五粮液、工商银行、科大讯飞3只个股的推广中保持着同样稳定精确的预测效果。

关键词: 股票预测, 投资者情绪, 新闻媒体信息, 信号分解, 门控单元

Abstract: Accurate prediction of future returns and risks in the stock market not only helps rational investors to invest more reasonably and effectively,but also provides useful guidance for policy makers and investors.Applying financial news headlines,this paper constructs an investor sentiment index that takes into account the cumulative effects of news using text analysis methods such as word embedding and machine learning.The Shanghai Composite Index is used as an example,and the empirical analysis decomposes the index’s fluctuation data into various inherent modes using the variational mode decomposition(VMD) method.Finally,the bidirectional gated recurrent unit(BiGRU) is introduced as a deep learning model for price fluctuation prediction.The results show that the investor sentiment index significantly affects the fluctuation of the Shanghai Composite index,and the influence of positive emotions and negative emotions is asymmetric.Considering the investor sentiment indicators and conducting the signal decomposition can effectively improve the prediction performance of stocks,and improve the prediction effect by up to 20%.In the benchmark scenario,the performance of VMD-BiGRU models is better than that of multiple econometric models and machine learning models,with higher accuracy and effectiveness,and the general performance of yield and volatility prediction is improved by more than 40%.The performance of model promotion in three stocks,Wuliangye,Industrial and Commercial Bank of China and IFLYTEK,maintain the same stable and accurate prediction effect.

Key words: Stock prediction, Investor sentiment, News media information, Signal decomposition, Gating unit

中图分类号:

F830.59

刘广, 易鸿. 融合媒体信息和信号分解的股票市场深度学习预测[J]. 计算机科学, 2024, 51(6A): 230600102-12. https://doi.org/10.11896/jsjkx.230600102

LIU Guang, YI Hong. Deep Learning Prediction of Stock Market Combining Media Information and Signal Decomposition[J]. Computer Science, 2024, 51(6A): 230600102-12. https://doi.org/10.11896/jsjkx.230600102

参考文献

[1]WANG W B,FEI L S,YI X M.China Stock Market Forecast based on EMD and Neural Network[J].Theory and Practice of System Engineering,2010,30(6):1027-1033.
[2]TIAN L H,WANG G Y,TAN D K.Reverse effect and asset pricing:How historical yields affect the present[J].Financial Research,2014(10):177-192.
[3]SONG G,ZHANG Y F,BAO F X,et al.Stock prediction model based on particle swarm optimization of LSTM[J].Journal of Beijing University of Aeronautics,2019,45(12):2533-2542.
[4]ZHAO Y X,VINEET U,CAI Y Z.Stock returns,quantile autocorrelation,and volatility forecasting[J].International Review of Financial Analysis,2021,73:101599.
[5]TAKAHASHI M,WATANABE T,OMORI Y.ForecastingDaily Volatility of Stock Price Index Using Daily Returns and Realized Volatility[J/OL].Econometrics and Statistics,2021.https://doi.org/10.1016/j.ecosta.2021.08.002.
[6]GHOSH P,NEUFELD A,SAHOO J.Forecasting directionalmovements of stock prices for intraday trading using LSTM and random forests[J].Finance Research Letters,2022,46:102280.
[7]ABU-MOSTAFA Y S,ATIYAA F.Introduction to financialforecasting[J].Applied intelligence,1996,6:205-213.
[8]HAMILTON J D.Time series analysis[M].Princeton:Princeton University Press,1994.
[9]ZHONG X,ENKE D.Forecasting daily stock market returnusing dimensionality reduction[J].Expert Systems with Applications,2017,67:126-139.
[10]TETLOC K,PAUL C.Giving content to investor sentiment:The role of media in the stock market[J].The Journal of Finance,2007,62(3):1139-1168.
[11]MOHAN S,MULLAPUDI S,SAMMETA S,et al.Stock price prediction using news sentiment analysis[C]//2019 IEEE Fifth International Conference on Big Data Computing Serviceand Applications(BigDataService).IEEE,2019:205-208.
[12]HUI B,LI P.Does investor sentiment predict stock returns?The evidence from Chinese stock market[J].Journal of Systems Science and Complexity,2014,27(1):130-143.
[13]ALANYALI M,MOAT H,PREIS T.Quantifying the Relationship Between Financial News and the Stock Market[J].Sci Rep,2013,3,3578.
[14]YU C M,GONG Y T,WANG F,et al.Stock trend prediction based on a text price fusion model[J].Data analysis and Knowledge Discovery,2018,2(12):33-42.
[15]NJOLSTAD P C S,HOYSAETER L S,WEI W,et al.Evaluating Feature Sets and Classifiers for Sentiment Analysis of Financial News[C]// IEEE/WIC/ACM International Joint Conferences on Web Intelligence & Intelligent Agent Technologies.ACM,2014:71-78.
[16]DAY M Y,LEE C C.Deep learning for financial sentiment analysis on finance news providers[C]// IEEE/ACM International Conference on Advances in Social Networks Analysis & Mining.ACM,2016.
[17]BARTOV E,LUCILE F,PARTHA M S.Can Twitter HelpPredict Firm-Level Earnings and Stock Returns?[J].Accounting REVIEW,2018,93(3):25-57.
[18]WU G R,HOU C T,LIN J L.Can economic news predict Taiwan stock market returns?[J].Asia Pacific Management Review,2019,24(1):54-59.
[19]LI X D,WU P J,WANG W P.Incorporating stock prices and news sentiments for stock market prediction[J].A case of Hong Kong.Information Processing & Management,2020,57(5):102212.
[20]AUDRINO F,SIGRIST F,BALLINARI D.The impact of sentiment and attention measures on stock market volatility[J].International Journal of Forecasting,2020,36(2):334-357.
[21]WAN X,YANG J,MARINOV S,et al.Sentiment correlation in financial news networks and associated market movements[J].Scientific Reports,2021,11:3062.
[22]WU S,LIU Y,ZOU Z,et al.S_I_LSTM:stock price prediction based on multiple data sources and sentiment analysis[J].Connection Science,2022,34(1):44-62.
[23]CAO J,LI Z,LI J.Financial time series forecasting model based on ceemdan and lstm[J].PHysica a-statistical Mechanics and ITS Applications,2019:0378-4371.
[24]MOGHA R,ADI L,HAMICHE M.Stock market predictionusing LSTM recurrent neural network[J].Procedia Computer Science,2020(170):1168-1173.
[25]SUNN Y,MD ARIF I,MIRZA MOHD SHAHRIAR M,et al.Deep learning-based stock price prediction using LSTM and bi-directional LSTM model[C]//2020 2nd Novel Intelligent and Leading Emerging Sciences Conference(NILES).IEEE,2020.
[26]BHAVANI A,RAMANA A V,CHAKRAVARTHY A S N.Comparative Analysis between LSTM and GRU in Stock Price Prediction[C]//2022 International Conference on Edge Computing and Applications(ICECAA).IEEE,2022:532-537.
[27]ANDERSEN T G,BOLLERSLEV T,DIEBOLD F X,et al.The distribution of realized stock return volatility[J].Journal of Financial Economics,2001,61(1):43-76.
[28]YANG X.Skewed Logistic distribution of log yields with theVaR estimate[J].Mathematical Statistics and Administration,2011,30(3):548-553.
[29]GAO W,XU F,ZHOU Z H.Towards convergence rate analysis of random forests for classification[J].Artificial Intelligence,2022,313:103788.
[30]BROWN G W,CLIFF MICHAEL T.Investor sentiment and the near-term stock market[J].Journal of Empirical Finance,2004,11(1):1-27.
[31]TETLOCK P C,SAAR-TSECHANSKY M,MACSKASSYS.More Than Words:Quantifying Language to Measure Firms’ Fundamentals[J].Journal of Finance,2008,63(3):1437-1467.
[32]DING T,FU X J,LIU M W.Wind electric power prediction based on the GA-VMD-BiLSTM algorithm[J].Journal of Yangzhou University(Natural Science Edition),2022,25(4):44-49.
[33]FAMA E F,FRENCH K R.Size and book-to-market factors in earnings and returns[J].The Journal of Finance,1995,50(1):131-155.
[34]MACKINLAY A C.Event Studies in Economics and Finance[J].Journal of Economic Literature,1997,35(1):13-39.
[35]DIEBOLD F X,MARIANO RS.Comparing predictive accuracy(Reprinted)[J].Journal of Business & Economic Statistics,1994,20(169).
[36]ROSSI B.Chapter 21-Advances in Forecasting under Instability[M]//ELLIOTT G,TIMMERMANN A.Handbook of Economic Forecasting:Vol.2.Elsevier,2013:1203-1324.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed