IMGAF-RLNet模型的股指趋势预测研究

doi:10.3778/j.issn.1002-8331.2311-0126

摘要/Abstract

摘要： 针对金融时间序列动态不稳定性以及长期依赖的特性，构建了基于深度学习算法的IMGAF-RLNet模型预测中国股票市场的大、中盘指数涨跌趋势。IMGAF-RLNet采用格拉姆角场方法将目标股指和基于斯皮尔曼秩相关系数筛选的成分股的不同特征序列编码为格拉姆差角场矩阵，然后将得到的矩阵序列构造为多维张量输入根据预训练模型分类结果筛选的CNN分类器残差网络（ResNet）进行特征提取，同时添加长短时记忆网络（LSTM）学习股指数据的时序特征，最后通过全连接网络对ResNet提取的局部特征和LSTM提取的整体特征完成股指趋势分类预测。选取沪深300、上证50、中证500指数作为研究对象。实验表明，三只股指的短、中、长期趋势预测准确率均在59%以上，其中预测效果最好的窗口及分类准确率分别为40、20、20以及62.65%、63.68%、61.85%。

关键词: 股指趋势预测, 数据增强, 格拉姆角场, 残差神经网络, 长短时记忆网络

Abstract: Aiming at the dynamic instability and long-term dependence of financial time series, an IMGAF-RLNet model based on deep learning algorithm is constructed to predict the rise and fall trends of large and medium cap indices in the Chinese stock market. IMGAF-RLNet uses the Gramian angular field method to encode the different feature sequences of the target stock index and the constituent stocks based on Spearman rank correlation coefficient screening into the Gramian difference angle field matrix. Then, the obtained matrix sequence is constructed as a multi-dimensional tensor input to a CNN classifier residual network (ResNet) screened based on the classification results of the pre-trained model for feature extraction, and a long short term memory network (LSTM) is added to learn the temporal features of the stock index data, Finally, the local features extracted by ResNet and the overall features extracted by LSTM are used to complete the classification and prediction of stock index trends through a fully connected network. The CSI 300, SSE 50, and CSI 500 indices are selected as the research subjects. The experiment shows that the accuracy of short-term, medium, and long-term trend prediction for the three stock indices is above 59%, with the best prediction window and classification accuracy are 40, 20, 20 and 62.65%, 63.68%, 61.85% respectively.

Key words: stock index trend prediction, data augmentation, Gramian angular field, residual network, long short term memory network (LSTM)

张菊平, 李路. IMGAF-RLNet模型的股指趋势预测研究[J]. 计算机工程与应用, 2025, 61(6): 229-243.

ZHANG Juping, LI Lu. IMGAF-RLNet Model for Stock Index Trend Forecasting[J]. Computer Engineering and Applications, 2025, 61(6): 229-243.

参考文献

[1] SHAHI T B, SHRESTHA A, NEUPANE A, et al. Stock price forecasting with deep learning: a comparative study[J]. Mathematics, 2020, 8(9): 1441.
[2] ZHAO J H, ZENG D L, LIANG S, et al. Prediction model for stock price trend based on recurrent neural network[J]. Journal of Ambient Intelligence and Humanized Computing, 2021, 12：745-753.
[3] ZENG X H, CAI J P, LIANG C Z, et al. A hybrid model integrating long short-term memory with adaptive genetic algorithm based on individual ranking for stock index prediction[J]. Plos One, 2022, 17(8): e0272637.
[4]?KHALIL A B, BAKAR A B U. A comparative study of deep learning algorithms in univariate and multivariate forecasting of the malaysian stock market[J]. Sains Malaysiana, 2023, 52(3): 993-1009.
[5] ZHAO Y L, YANG G. Deep learning-based integrated framework for stock price movement prediction[J]. Applied Soft Computing, 2023, 133: 109921.
[6] DI-GIORGI G, SALAS R, AVARIA R, et al. Volatility forecasting using deep recurrent neural networks as GARCH models?[J]. Computational Statistics, 2023, 36(1): 1377-1401.
[7] JARRAH M, DERBALI M. Predicting saudi stock market index by using multivariate time series based on deep learning[J]. Applied Sciences, 2023, 13(14): 8356.
[8] SEZER O B, OZBAYOGLU A M. Algorithmic financial trading with deep convolutional neural networks: time series to image conversion approach[J]. Applied Soft Computing, 2018, 70: 525-538.
[9] WEN M, LI P, ZHANG L F, et al. Stock market trend prediction using high-order information of time series[J]. IEEE Access, 2019: 28299-28308.
[10] SEZER B O, OZBAYOGLU M A. Financial trading model with stock bar chart image time series with deep convolutional neural networks[J]. arXiv:1903.04610, 2019.
[11] BARRA S, CARTA A, CORRIGA A, et al. Deep learning and time series-to-image encoding for financial forecasting[J]. IEEE/CAA Journal of Automatica Sinica, 2020, 7(3): 683-692.
[12] CHEN W, JIANG M, ZHANG W G, et al. A novel graph convolutional feature based convolutional neural network for stock trend prediction[J]. Information Sciences, 2021, 556: 67-94.
[13] HUNG C C, CHEN?Y J. DPP: deep predictor for price movement from candlestick charts[J]. Plos One, 2021, 16(6): e0252404.
[14] AKSEHIR Z D, KILIC E. How to handle data imbalance and feature selection problems in CNN-based stock price forecasting[J]. IEEE Access, 2022, 10: 31297-31305.
[15] GHASEMIEH A, KASHEF R. A robust deep learning model for predicting the trend of stock market prices during market crash periods[C]//Proceedings of the 2022 IEEE International Systems Conference, 2022: 1-8.
[16] CHANDAR S K. Convolutional neural network for stock trading using technical indicators[J]. Automated Software Engineering, 2022, 29: 1-14.
[17] CHENN Y. Visual recognition and prediction analysis of China’s real estate index and stock trend based on CNN-LSTM algorithm optimized by neural networks[J]. Plos One, 2023, 18(2): e0282159.
[18] MA D, YUAN D, HUANG M J, et al. VGC-GAN: a multi-graph convolution adversarial network for stock price prediction[J]. Expert Systems with Applications, 2024, 236: 121204.
[19] ZHANG X S, MA Y L, WANG M H. An attention-based Logistic-CNN-BiLSTM hybrid neural network for credit risk prediction of listed real estate enterprises[J]. Expert Systems, 2024, 41(2): e13299.
[20] ZHANG J L, YE L S, LAI Y Z. Stock price prediction using CNN-BiLSTM-Attention model[J]. Mathematics, 2023, 11(9): 1985.
[21] WANG Z G, OATES T. Encoding time series as images for visual inspection and classification using tiled convolutional neural networks[C]//Proceedings of the the Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015.
[22] SCHMIDHUBER J, HOCHREITER S. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.
[23] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[24] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[25] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J]. arXiv:1409. 1556, 2014.
[26] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015: 1-9.
[27] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[28] SONG H, CHOI H. Forecasting stock market indices using the recurrent neural network based hybrid models: CNN-LSTM, GRU-CNN, and ensemble models[J]. Applied Sciences, 2023, 13(7): 4644.
[29] YAO Y, ZHANG Z Y, ZHAO Y. Stock index forecasting based on multivariate empirical mode decomposition and temporal convolutional networks[J]. Applied Soft Computing, 2023, 142: 110356.

编辑推荐 0

Metrics

阅读次数

全文

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	0	0	0	28

	来源	本网站

	次数	28
	比例	100%

摘要

最新录用	在线预览	正式出版

0	0	25

	来源	本网站

	次数	25
	比例	100%