基于分解-集成和混频数据采样的中国股票市场预测研究

陈凯杰, 唐振鹏, 吴俊传, 杜晓旭, 蔡毅

系统工程理论与实践 ›› 2022, Vol. 42 ›› Issue (11) : 3105-3120.

PDF(2033 KB)
中国科学院数学与系统科学研究院期刊网
PDF(2033 KB)
系统工程理论与实践 ›› 2022, Vol. 42 ›› Issue (11) : 3105-3120. DOI: 10.12011/SETP2022-0376
论文

基于分解-集成和混频数据采样的中国股票市场预测研究

    陈凯杰1, 唐振鹏2, 吴俊传3, 杜晓旭1, 蔡毅1
作者信息 +

Forecasting China's stock index: A hybrid method based on decomposition-integrated and mixed-frequency data

    CHEN Kaijie1, TANG Zhenpeng2, WU Junchuan3, DU Xiaoxu1, CAI Yi1
Author information +
文章历史 +

摘要

股票市场的准确预测对投资者和监管层而言都具有重要意义.在金融时间序列预测研究中分解-集成框架被广泛使用,然而以往研究中针对分解后的模态分量多数采用单一历史数据预测未来的思路,往往忽略了其他低频异质数据对分量的影响.本文融合了分解-集成与混频数据采样思想,提出EEMD-Mixed Frequency CNN-BiLSTM-Attention/LSTM-LSTM (EE-MFCBA/L-L)股指收益率预测模型,通过EEMD将股指收益率分解为若干不同频率特征的分量,采用模糊熵算法识别分量频率特征,进而结合不同频率倍差的低频数据,使用MFCBA/L模型实现对模态分量的预测,最后采用LSTM模型非线性集成各分量的预测结果.实证结果表明,所提出的模型可以更好地适应收益率变化特征,与传统模型相比,所提模型在预测非平稳和非线性收益率序列方面具有显著优势,具有最低预测误差和最高的方向性预测准确率.

Abstract

Accurate forecasts of the stock market have important implications for both investors and regulators. The decomposition-integration framework is widely used in forecasting research of financial time series. However, most of previous studies only use single historical data to predict the components, which ignoring the influence of other low-frequency heterogeneous data on the components. This study proposes a novel decomposition-integration model EEMD-Mixed Frequency CNN-BiLSTM-Attention/LSTM-LSTM (EE-MFCBA/L-L), which take the advantages of the decomposition-integration and mixing data sampling. Firstlly the stock index is decomposed into several different frequency components by EEMD. Meanwhile the fuzzy entropy algorithm is used to identify the frequency characteristic of the components. Then the components are predicted by MFCBA/L model where the low frequency data will be considered according to the frequency characteristics of the components. Finally, the LSTM model is used to nonlinearly integrate the predictor of each component. The empirical results show that the proposed model can better adapt to the characteristics of returns. Compared with the traditional model, the proposed model has significant advantages in predicting non-stationary and nonlinear return series, with the lowest prediction error and the highest directional prediction accuracy.

关键词

股票指数 / 混频预测 / 集成学习 / 信号分解

Key words

stock index / mixed frequency prediction / ensemble learning / signal decomposition

引用本文

EndNote

Ris (Procite)

Bibtex

导出引用
陈凯杰 , 唐振鹏 , 吴俊传 , 杜晓旭 , 蔡毅. 基于分解-集成和混频数据采样的中国股票市场预测研究. 系统工程理论与实践, 2022, 42(11): 3105-3120 https://doi.org/10.12011/SETP2022-0376
CHEN Kaijie , TANG Zhenpeng , WU Junchuan , DU Xiaoxu , CAI Yi. Forecasting China's stock index: A hybrid method based on decomposition-integrated and mixed-frequency data. Systems Engineering - Theory & Practice, 2022, 42(11): 3105-3120 https://doi.org/10.12011/SETP2022-0376
中图分类号： F830.9    F224   

参考文献

[1] 吴俊传, 唐振鹏, 杜晓旭, 等. 基于改进的对数周期幂律模型研究中国股市崩盘预警[J]. 物理学报, 2022, 71(2): 9-18.Wu J C, Tang Z P, Du X X, et al. Research on Chinese stock market crash early warning based on improved log-periodic power law model[J]. Acta Physica Sinica, 2022, 71(2): 9-18.
[2] Abu-Mostafa Y S, Atiya A F. Introduction to financial forecasting[J]. Applied intelligence, 1996, 6(3): 205-213.
[3] Kocak C. ARMA (p, q) type high order fuzzy time series forecast method based on fuzzy logic relations[J]. Applied Soft Computing, 2017, 58: 92-103.
[4] Adebiyi A A, Adewumi A O, Ayo C K. Comparison of ARIMA and artificial neural networks models for stock price prediction[J]. Journal of Applied Mathematics, 2014. https://doi.org/10.1155/2014/614342.
[5] Lin Y, Xiao Y, Li F. Forecasting crude oil price volatility via a HM-EGARCH model[J]. Energy Economics, 2020, 87: 104693.
[6] Chen Y, Hao Y. A feature weighted support vector machine and K-nearest neighbor algorithm for stock market indices prediction[J]. Expert Systems with Applications, 2017, 80: 340-355.
[7] Oliveira V, Sousa V, Dias-Ferreira C. Artificial neural network modelling of the amount of separately-collected household packaging waste[J]. Journal of Cleaner Production, 2019, 210: 401-409.
[8] Davis P, Aziz F, Newaz M T, et al. The classification of construction waste material using a deep convolutional neural network[J]. Automation in Construction, 2021, 122: 103481.
[9] Yuan X, Li L, Wang Y. Nonlinear dynamic soft sensor modeling with supervised long short-term memory network[J]. IEEE Transactions on Industrial Informatics, 2019, 16(5): 3168-3176.
[10] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.
[11] Zhao X, Li P, Xiao K, et al. Sensor drift compensation based on the improved LSTM and SVM multi-class ensemble learning models[J]. Sensors, 2019, 19(18): 3844.
[12] Wang W, Shen J. Deep visual attention prediction[J]. IEEE Transactions on Image Processing, 2017, 27(5): 2368-2378.
[13] Long W, Lu Z, Cui L. Deep learning-based feature engineering for stock price movement prediction[J]. Knowledge-Based Systems, 2019, 164: 163-173.
[14] Zhou Z, Lin L, Li S. International stock market contagion: A CEEMDAN wavelet analysis[J]. Economic Modelling, 2018, 72: 333-352.
[15] Cheng C H, Wei L Y. A novel time-series model based on empirical mode decomposition for forecasting TAIEX[J]. Economic Modelling, 2014, 36: 136-141.
[16] Yu L, Wang S, Lai K K. A neural-network-based nonlinear metamodeling approach to financial time series forecasting[J]. Applied Soft Computing, 2009, 9(2): 563-574.
[17] Li Y, Wang S, Wei Y, et al. A new hybrid VMD-ICSS-BiGRU approach for gold futures price forecasting and algorithmic trading[J]. IEEE Transactions on Computational Social Systems, 2021, 8(6): 1357-1368.
[18] Huang N E, Shen Z, Long S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society of London, Series A: Mathematical, Physical and Engineering Sciences, 1998, 454(1971): 903-995.
[19] 唐振鹏, 吴俊传, 张婷婷, 等. 基于二次分解和集成学习的粮食期货价格预测研究[J]. 系统工程理论与实践, 2021, 41(11): 2837-2849. Tang Z P, Wu J C, Zhang T T, et al. Research on grain futures price forecasting based on secondary decomposition and ensemble learning[J]. Systems Engineering — Theory & Practice, 2021, 41(11): 2837-2849.
[20] 孙少龙, 魏云捷, 汪寿阳. 基于分解-聚类-集成学习的汇率预测方法[J]. 系统工程理论与实践, 2022, 42(3): 664-677.Sun S L, Wei Y J, Wang S Y. Exchange rates forecasting with decomposition-clustering-ensemble learning approach[J]. Systems Engineering — Theory & Practice, 2022, 42(3): 664-677.
[21] Wang T, Zhang M, Yu Q, et al. Comparing the applications of EMD and EEMD on time-frequency analysis of seismic signal[J]. Journal of Applied Geophysics, 2012, 83: 29-34.
[22] Girardin E, Joyeux R. Macro fundamentals as a source of stock market volatility in China: A GARCH-MIDAS approach[J]. Economic Modelling, 2013, 34: 59-68.
[23] 林忠国, 韩立岩. 机构交易的正面效应: 波动率和市场效率的视角[J]. 系统工程理论与实践, 2011, 31(4): 606-616. Lin Z G, Han L Y. Positive influence of institutional trading: The perspectives of volatility and market efficiency[J]. Systems Engineering — Theory & Practice, 2011, 31(4): 606-616.
[24] 胡大春, 金赛男. 基金持股比例与A股市场收益波动率的实证分析[J]. 金融研究, 2007(4): 129-142. Hu D C, Jin S N. An empirical analysis of institutional holding and A-share market return volatility[J]. Journal of Financial Research, 2007(4): 129-142.
[25] Wu Z, Huang N E. Ensemble empirical mode decomposition: A noise-assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2009, 1(1): 1-41.
[26] Ortiz-Cruz A, Rodriguez E, Ibarra-Valdez C, et al. Efficiency of crude oil markets: Evidences from informational entropy analysis[J]. Energy Policy, 2012, 41: 365-373.
[27] Wang K, Fu W, Chen T, et al. A compound framework for wind speed forecasting based on comprehensive feature selection, quantile regression incorporated into convolutional simplified long short-term memory network and residual error correction[J]. Energy Conversion and Management, 2020, 222: 113234.
[28] Gu J, Wang Z, Kuen J, et al. Recent advances in convolutional neural networks[J]. Pattern Recognition, 2018, 77: 354-377.
[29] Wu Y X, Wu Q B, Zhu J Q. Improved EEMD-based crude oil price forecasting using LSTM networks[J]. Physica A: Statistical Mechanics and its Applications, 2019, 516: 114-124.
[30] Wang Q, Dai X, Zhou D. Dynamic correlation and risk contagion between ''black'' futures in China: A multi-scale variational mode decomposition approach[J]. Computational Economics, 2020, 55(4): 1117-1150.
[31] Lu W, Li J, Wang J, et al. A CNN-BiLSTM-AM method for stock price prediction[J]. Neural Computing and Applications, 2021, 33(10): 4741-4753.
[32] Wang H, Wang J, Cao L, et al. A stock closing price prediction model based on CNN-BiSLSTM[J]. Complexity, 2021, 2021.
[33] Azami H, Li P, Arnold S E, et al. Fuzzy entropy metrics for the analysis of biomedical signals: Assessment and comparison[J]. IEEE Access, 2019, 7: 104833-104847.
[34] Hansen P R, Lunde A, Nason J M. The model confidence set[J]. Econometrica, 2011, 79(2): 453-497.
[35] Wang Y, Ma F, Wei Y, et al. Forecasting realized volatility in a changing world: A dynamic model averaging approach[J]. Journal of Banking & Finance, 2016, 64: 136-149.
[36] Diebold F X, Mariano R S. Comparing predictive accuracy[J]. Journal of Business & Economic Statistics, 2002, 20(1): 134-144.
[37] 余秋玲, 朱宏泉. 宏观经济信息与股价联动——基于中国市场的实证分析[J]. 管理科学学报, 2014, 17(3): 15-26.Yu Q L, Zhu H Q. Macroeconomic information and stock price synchronicity: Empirical analysis in Chinese stock markets[J]. Journal of Management Sciences in China, 2014, 17(3): 15-26.
[38] 寇明婷, 杨海珍, 汪寿阳. 股票价格与宏观经济联动关系研究——政策预期视角[J]. 管理评论, 2018, 30(9): 3-11.Kou M T, Yang H Z, Wang S Y. Analysis of the correlation between stock market and macroeconomic operation: A perspective of policy expectation[J]. Management Review, 2018, 30(9): 3-11.
[39] 刘洋, 谢栌乐, 汪寿阳, 等. 基于 EEMD-AWNN 集成学习的中国经常账户预测研究[J]. 系统工程理论与实践, 2021, 41(5): 1240-1251.Liu Y, Xie L Y, Wang S Y, et al. Forecasting China's current account with EEMD-AWNN ensemble learning approach[J]. Systems Engineering — Theory & Practice, 2021, 41(5): 1240-1251.

基金

国家自然科学基金面上项目(71973028, 71573042)
PDF(2033 KB)

831

Accesses

0

Citation

Detail
段落导航
相关文章

/

〈 〉