A Heterogeneous Multi-Task Deep Learning Approach for Geopolitical Risk Prediction and Cross-Country Heterogeneity in ASEAN Equity Markets
DOI:
https://doi.org/10.62411/jcta.17610Keywords:
ASEAN Equity Markets , Cross-Country Heterogeneity, Deep Learning, Emerging Financial Markets, Explainable AI (SHAP), Geopolitical Risk Prediction, Heterogeneous Multi-Task Learning, Volatility ForecastingAbstract
Modeling the asymmetric transmission of geopolitical shocks across structurally diverse emerging markets presents a severe computational challenge. Traditional homogeneous deep learning approaches often suffer from negative transfer and base-rate bias when applied to cross-country financial prediction. To address this, we propose the Heterogeneous Multi-Task Spatiotemporal Network (HMT-STN), a structurally adaptive framework that integrates country-specific embeddings, heterogeneous task decoupling, and a Focal Loss mechanism. Evaluated on a daily panel dataset of ASEAN-4 equity markets (2016–2026), the HMT-STN is designed to simultaneously optimize directional return classification and volatility regression. Comprehensive evaluations using Precision, Recall, F1-score, and AUC-ROC demonstrate the framework's superior discriminative power, notably achieving 66.21% accuracy and an AUC-ROC of 0.68 in Thailand, a highly volatile transitional market. Rigorous ablation studies confirm that Focal Loss prevents catastrophic performance collapse in noisy environments, while country embeddings successfully isolate market-specific idiosyncrasies. Furthermore, SHAP (SHapley Additive exPlanations) analysis reveals a tripartite spectrum of cross-country heterogeneity, providing model-based feature attributions that are consistent with established economic theories of market segmentation and geopolitical risk sensitivity. Collectively, this study demonstrates that explicitly modeling cross-country heterogeneity through heterogeneous multi-task learning offers a robust, interpretable, and computationally efficient paradigm for geopolitical risk assessment in fragmented global economies.References
P. Basu and P. Ray, “China-plus-one: expanding global value chains,” J. Bus. Strategy, vol. 43, no. 6, pp. 350–356, 2022, doi: 10.1108/JBS-04-2021-0066.
L. Han, D. Ning, and S. Xin, “Geopolitical risk and economic growth: Evidence from ASEAN countries,” Financ. Res. Lett., vol. 88, p. 109102, 2026, doi: 10.1016/j.frl.2025.109102.
K. Tissaoui and N. Azibi, “Sentiment momentum and regime-dependent predictability in crude oil markets: A Markov Switching analysis with higher-order VIX and MOVE derivatives,” Int. Rev. Econ. & Financ., vol. 108, p. 105197, 2026, doi: 10.1016/j.iref.2026.105197.
Z. Sun, “A Comprehensive Review of Stock Index Prediction Methods: From Traditional Econometrics to Deep Learning with Attention Mechanisms,” Trans. Comput. Sci. Intell. Syst. Res., vol. 9, pp. 547–552, 2025, doi: 10.62051/g34zas22.
K. Gerakoudi, D. Georgoulas, and P. J. Stavroulakis, “A novel hybrid ARIMA-LSTM model for maritime shipping stock forecasting: comparative evidence against statistical and machine learning benchmarks,” Marit. Bus. Rev., vol. 11, no. 1, pp. 2–20, 2026, doi: 10.1108/mabr-04-2024-0035.
X. Jiang, “A Multi-Task Learning Model for Stock Market Forecasting,” in Proceedings of SEML 2025 Symposium: Machine Learning Theory and Applications, 2025, pp. 1–8. doi: 10.54254/2755-2721/2025.TJ21873.
S. Vandenhende, S. Georgoulis, W. Van Gansbeke, M. Proesmans, D. Dai, and L. Van Gool, “Multi-Task Learning for Dense Prediction Tasks: A Survey,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 7, pp. 3614–3633, 2022, doi: 10.1109/TPAMI.2021.3054719.
O. B. Sezer, M. U. Gudelek, and A. M. Ozbayoglu, “Financial time series forecasting with deep learning: A systematic literature review: 2005--2019,” Appl. Soft Comput., vol. 90, p. 106181, 2020, doi: 10.1016/j.asoc.2020.106181.
T. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “Focal Loss for Dense Object Detection,” IEEE Trans. Pattern Anal. Mach. Intell., pp. 318–327, 2020, doi: 10.1109/TPAMI.2018.2858826.
G. Zou, Z. Lai, T. Wang, Z. Liu, and Y. Li, “MT-STNet: A Novel Multi-Task Spatiotemporal Network for Highway Traffic Flow Prediction,” IEEE Trans. Intell. Transp. Syst., pp. 8221–8236, 2024, doi: 10.1109/TITS.2024.3411638.
F. Boabang and S. A. Gyamerah, “An Enhanced Focal Loss Function to Mitigate Class Imbalance in Auto Insurance Fraud Detection with Explainable {AI},” 2025. doi: 10.48550/arXiv.2508.02283.
K.-H. Wang, Z.-S. Wang, H.-W. Liu, and X. Li, “Economic policy uncertainty and geopolitical risk: evidence from China and Southeast Asia,” Asia. Pac. Econ. Lit., vol. 37, no. 2, 2023, doi: 10.1111/apel.12388.
D. Mariyani, H. Hariyanti, and A. Kalista, “Shifting The Relationship Between Market Sentiment, Market Volume, Trading Volume and Volatility on The Indonesia Stock Exchange Approach Model Vector Autoregression,” Eduvest -- J. Univers. Stud., vol. 5, no. 4, pp. 3832–3842, 2025, doi: 10.59188/eduvest.v5i4.34659.
S. Chancharat and N. Chancharat, “Asymmetric spillover and quantile linkage between the United States and ASEAN + 6 stock returns under uncertainty,” J. Open Innov. Technol. Mark. Complex., vol. 10, no. 3, p. 100317, 2024, doi: 10.1016/j.joitmc.2024.100317.
R. D. Hartanto, G. F. Shidik, F. Alzami, A. Z. Fanani, and A. Marjuni, “Attention-Augmented {GRU} for Stock Forecasting: A Trade-Off Between Directional Accuracy and Price Prediction Error,” J. Comput. Theor. Appl., 2026, doi: 10.62411/jcta.15863.
P. O. Odion, M. M. Lawal, and A. Abdulrauf, “A Comparative Analysis of an Enhanced Hybrid Model for Predicting Dollar Against Naira Exchange Rate Using Deep Learning and Statistical Methods,” J. Comput. Theor. Appl., 2025, doi: 10.62411/jcta.12513.
J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, “Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling,” 2014. doi: 10.48550/arXiv.1412.3555.
M. Ahmed, A. A. Nova, R. Pervez, M. N. Nagib, J. Maua, and M. F. Mridha, “RiskAwareTNet: A Multi-Task Deep Learning Framework for Stock Price Prediction and Risk Assessment,” in 2025 International Conference on Electrical, Computer and Communication Engineering (ECCE), IEEE, 2025, pp. 1–6. doi: 10.1109/ECCE64574.2025.11013888.
X. Zhao, Y. Liu, and Q. Zhao, “Generalized Loss-Based CNN-BiLSTM for Stock Market Prediction,” Int. J. Financ. Stud., vol. 12, no. 61, pp. 1–23, 2024, doi: 10.3390/ijfs12030061.
X. Zhao, “False Awareness Stock Market Prediction by LightGBM with Focal Loss,” in Proceedings of the International Conference on Research in Adaptive and Convergent Systems (RACS ’22), Aizu, Japan: Association for Computing Machinery, 2022, pp. 147–152. doi: 10.1145/3538641.3561502.
Y. A. Bipasha, “Market efficiency, anomalies and behavioral finance: A review of theories and empirical evidence,” World J. Adv. Res. Rev., pp. 827–839, 2022, doi: 10.30574/wjarr.2022.15.2.0876.
L. Feng, Y. Zheng, X. Wang, C. Guo, and R. Xue, “Global stock market forecasting: Insights from series and parallel combination of machine learning models,” Pacific-Basin Financ. J., vol. 93, p. 102905, 2025, doi: 10.1016/j.pacfin.2025.102905.
U. Razi, C. W. H. Cheong, S. Shams, T. Sarker, A. Sharif, and S. Afshan, “Assessing the turbulence: Wavelet coherence and causality analysis of energy price volatility and exchange rate instability,” Energy, vol. 331, p. 136948, 2025, doi: 10.1016/j.energy.2025.136948.
Z. Tan, M. Hu, B. Liu, and G. Yin, “Futures Quantitative Investment with Heterogeneous Continual Graph Neural Network,” in 2024 IEEE International Conference on Data Mining (ICDM), IEEE, 2024, pp. 851–856. doi: 10.1109/ICDM59182.2024.00104.
A. Rohan, D. Hossen, and N. Pranto, “Artificial intelligence in financial market prediction: advancements in machine learning for stock price forecasting,” Front. Artif. Intell., vol. 8, pp. 1–34, 2026, doi: 10.3389/frai.2025.1696423.
S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” 2017, Curran Associates.
M. L. de Prado, Advances in Financial Machine Learning, 1st ed. Wiley, 2018.
D. Wang, L. Tian, and Y. Zhao, “Smoothed empirical likelihood for the Youden index,” Comput. Stat. & Data Anal., vol. 115, pp. 1–10, 2017, doi: 10.1016/j.csda.2017.03.014.
G. G. Goswami, M. Yahya, and M. Aftabi, “Impact of financial and energy market uncertainties on ASEAN-5 markets,” Eurasian Econ. Rev., vol. 15, pp. 1261–1283, 2025, doi: 10.1007/s40822-025-00327-w.
K. Y. Yeoh and K. Tham, “Exploring the relationship between macroeconomic conditions and distressed real estate loans during the COVID-19 pandemic in Malaysia and Singapore,” Int. J. Hous. Mark. Anal., pp. 1–40, 2025, doi: 10.1108/IJHMA-04-2025-0100.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 St. Dwiarso Utomo, Entot Suhartono, Ngurah Pandji M. A. D., Bambang Minarso, Agung Prajanto, Shujahat Ali
This work is licensed under a Creative Commons Attribution 4.0 International License.
