Abstract
We present a stock index prediction model based on a multi-time scale learning (MTSL) and the multi-graph attention network (MGAT) approach. Instead of dealing with individual stock markets, we consider a group of stock markets simultaneously and exploit the effects of their interactions when making predictions. Our model consists of a boosting Hodrick-Prescott (bHP) filter and MGATs, along with MTSL processes. The bHP filter decomposes the stock index series into the slow-varying growth trend and the fast-varying cyclical volatility for training the MGAT to facilitate the learning of multi-time scale data. The MGAT exploits the interplays of stock markets with three different types of graphs: a regionality graph that qualitatively describes the linkages among stock markets within the same region or similar financial systems; a correlativity graph that quantifies the Pearson correlation between stock markets; and a causality graph that measures the convergence cross mapping (CCM) causality between stock markets. Particularly, the last graph is essentially a directed graph, which captures the nonreciprocal relationship between different stock markets. Experimental results on real stock indices reveal the effectiveness and merit of the proposed model in comparison with other models adopted in this paper.
Similar content being viewed by others
References
Xu Y, Yang C, Peng S, Nojima Y (2020) A hybrid two-stage financial stock forecasting algorithm based on clustering and ensemble learning. Appl Intell 50(11):3852–3867
Bazrkar MJ, Hosseini S (2022) Predict stock prices using supervised learning algorithms and particle swarm optimization algorithm. Comput Econ:1–22
Zhang J, Li L, Chen W (2021) Predicting stock price using two-stage machine learning techniques. Comput Econ 57(4):1237–1261
Wang X, Li X, Li S (2022) A novel stock indices hybrid forecasting system based on features extraction and multi-objective optimizer. Appl Intell:1–24
Jia L, Li W, Qiao J (2022) An online adjusting RBF neural network for nonlinear system modeling. Appl Intell:1–14
Singh R, Srivastava S (2017) Stock prediction using deep learning. Multimed Tools Appl 76 (18):18569–18584
Wang J, Wang J (2017) Forecasting stochastic neural network based on financial empirical mode decomposition. Neural Netw 90:8–20. https://doi.org/10.1016/j.neunet.2017.03.004
Niu H, Xu K, Wang W (2020) A hybrid stock price index forecasting model based on variational mode decomposition and LSTM network. Appl Intell 50(12):4296–4309
Sitte R, Sitte J (2002) Neural networks approach to the random walk dilemma of financial time series. Appl Intell 16(3):163–171
Karevan Z, Suykens JA (2020) Transductive LSTM for time-series prediction: an application to weather forecasting. Neural Netw 125:1–9
Koo E, Kim G (2022) A new neural network approach for predicting the volatility of stock market. Comput Econ:1–15
Khandelwal I, Adhikari R, Verma G (2015) Time series forecasting using hybrid ARIMA and ANN models based on DWT decomposition. Procedia Comput Sci 48:173–179. https://doi.org/10.1016/j.procs.2015.04.167
Chen W, Jiang M, Zhang WG, Chen Z (2021) A novel graph convolutional feature based convolutional neural network for stock trend prediction. Inf Sci 556:67–94
Cheng D, Yang F, Xiang S, Liu J (2022) Financial time series forecasting with multi-modality graph neural network. Pattern Recogn 121:108218
Phillips PCB, Shi Z (2021) Boosting: why you can use the HP filter. Int Econ Rev 62 (2):521–570
Sugihara G, May R, Ye H, Hsieh C-H, Deyle E, Fogarty M, Munch S (2012) Detecting causality in complex ecosystems. Science 338(6106):496–500
Stolbov M, Shchepeleva M (2020) Systemic risk, economic policy uncertainty and firm bankruptcies: Evidence from multivariate causal inference. Res Int Business Finance 52:101172. https://doi.org/10.1016/j.ribaf.2019.101172
Wu T, Gao X, An S, Liu S (2021) Time-varying pattern causality inference in global stock markets. Int Rev Financial Anal 77:101806. https://doi.org/10.1016/j.irfa.2021.101806
(2020). Yahoo Finance: stock data. [EB/OL]. https://hk.finance.yahoo.com/ Accessed 30 Dec 2020
Hodrick RJ, Prescott EC (1997) Postwar U.S. business cycles: an empirical investigation. J Money Credit Bank 29(1):1–16. https://doi.org/10.2307/2953682
Wang Y, Yang J, Chen Y, De Maeyer P, Li Z, Duan W (2018) Detecting the causal effect of soil moisture on precipitation using convergent cross mapping. Sci Reports 8(1):1–8
Dost F, Maier E (2018) E-commerce effects on energy consumption: a multi-year ecosystem-level assessment. J Ind Ecol 22(4):799–812
Velickovic P, Cucurull G, Casanova A, Romero A, Lio P, Bengio Y (2018) Graph attention networks. In: International conference on learning representations (ICLR), pp 1–12
Liu Q, Ghosh S, Li J, Wong L, Ramamohanarao K (2018) Discovering pan-correlation patterns from time course data sets by efficient mining algorithms. Computing 100(4):421– 437
Weron R, Zator M (2015) A note on using the Hodrick-Prescott filter in electricity markets. Energy Econ 48:1–6. https://doi.org/10.1016/j.eneco.2014.11.014
Das A (2016) Cyclical behavior analysis of indian market using HP filter and spectral techniques. IUP J Appl Finance 22(2):62–78
Weerakody PB, Wong KW, Wang G, Ela W (2021) A review of irregular time series data handling with gated recurrent neural networks. Neurocomputing 441:161–178
Kipf TN, Welling M (2017) Semi-supervised classification with graph convolutional networks. In: International conference on learning representations (ICLR)
Deng C, Huang Y, Hasan N, Bao Y (2022) Multi-step-ahead stock price index forecasting using long short-term memory model with multivariate empirical mode decomposition. Inf Sci:297–321
Banerjee T, Sinha S, Choudhury P (2022) Long term and short term forecasting of horticultural produce based on the LSTM network model. Appl Intell 52(8):9117–9147
Acknowledgements
Project supported by the National Natural Science Foundation of China (Grant Nos. 61673027 and 62106047), Beijing Natural Science Foundation (No. 19L2037)) and the Industry-University-Research Innovation Fund for Chinese Universities (No. 2020HYB04001).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Liu, Y., Zhang, Q. & Chu, T. Stock index prediction based on multi-time scale learning with multi-graph attention networks. Appl Intell 53, 16263–16274 (2023). https://doi.org/10.1007/s10489-022-04285-7
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-022-04285-7