A novel financial modeling framework is presented that dynamically quantifies geopolitical risk for Taiwan’s stock market under Cross-Strait tensions, addressing the limitations of traditional asset pricing models in capturing spatiotemporal risk propagation. The framework integrates a Geopolitical Risk Propagation Module (GRPM), which combines dynamic network theory and sequence-to-sequence learning to model both the spatial spillover effects of geopolitical shocks and their temporal evolution. The GRPM consists of two core components: a Dynamic Asset Network (DAN) that tracks real-time risk transmission across asset classes and an LSTM-Event Encoder (LEE) that processes geopolitical event sequences to generate context-aware risk scores. These components interact through a Risk Propagation Layer (RPL), which adjusts spillover intensities based on event severity and historical asset correlations, thereby capturing market overreaction phenomena. The output is a time-varying risk-adjusted covariance matrix that enhances conventional multifactor models by explicitly incorporating geopolitical risk. Key innovations include the coupling of temporal event sequencing with spatial risk diffusion, a self-attentive mechanism to isolate high-impact events, and nonlinear spillover adjustments that reflect empirical market behavior. Implemented with PyTorch Geometric and Hugging Face Transformers, the framework demonstrates practical applicability by ingesting real-time data from Bloomberg and GDELT. Our approach not only improves risk sensitivity in financial models but also provides policymakers and investors with a tool to anticipate market disruptions during geopolitical crises. The methodology is particularly relevant for regions exposed to volatile political dynamics, offering a scalable template for other emerging markets. 

Wenjin Yang. Dynamic Geopolitical Risk Quantification in Financial Models: A Hybrid LSTM-Network Approach for Cross-Strait Tensions. Academic Journal of Business & Management (2025), Vol. 7, Issue 10: 7-16. https://doi.org/10.25236/AJBM.2025.071002.

[1] Benninga, S. (2014). Financial modeling. books.google.com.

[2] Mazzarisi, P. (2019). Dynamic network models with applications to finance. ricerca.sns.it.

[3] Lindemann, B., Müller, T., Vietz, H., Jazdi, N., & Weyrich, M. (2021). A survey on long short-term memory networks for time series prediction. Procedia Cirp.

[4] Liu, J., Liu, X., Wu, W., & Zheng, W. (2025). The effect of geopolitical uncertainty on stock liquidity: Evidence from taiwan misfired missile event. Applied Economics.

[5] Focardi, S., & Fabozzi, F. (2004). The mathematics of financial modeling and investment management. books.google.com.

[6] Svetlova, E. (2012). On the performative power of financial models. Economy and Society.

[7] Wang, H. (2019). The causality link between political risk and stock prices: A counterfactual study in an emerging market. Journal of Financial Economic Policy.

[8] Beltratti, A., Margarita, S., & Terna, P. (1996). Neural networks for economic and financial modelling. jasss.soc.surrey.ac.uk.

[9] Wu, B., & Wang, Q. (2025). Cross-asset contagion and risk transmission in global financial networks. The North American Journal of Economics and Finance.

[10] Buraschi, A., & Tebaldi, C. (2024). Financial contagion in network economies and asset prices. Management Science.

[11] Barnett, W., Wang, X., Xu, H., & Zhou, W. (2022). Hierarchical contagions in the interdependent financial network. Journal of Financial Stability.

[12] Umar, Z., Bossman, A., Choi, S., & Vo, X. (2023). Information flow dynamics between geopolitical risk and major asset returns. PLoS One.

[13] Gong, X., Ning, H., & Xiong, X. (2025). Research on the cross-contagion between international stock markets and geopolitical risks: The two-layer network perspective. Financial Innovation.

[14] Jin, Q., Sun, L., Chen, Y., & Hu, Z. (2024). Financial risk contagion based on dynamic multi-layer network between banks and firms. Physica A: Statistical Mechanics and Its Applications.

[15] Deng, Y., & Wu, Y. (2024). Does geopolitical risk matter for cross‐industry risk contagion: The roles of real linkage and information channels. Asia‐Pacific Journal of Financial Studies.

[16] Bartesaghi, P., Benzi, M., Clemente, G., Grassi, R., et al. (2020). Risk-dependent centrality in economic and financial networks. SIAM Journal on Applied Mathematics.

[17] Zhang, Y. (2004). Prediction of financial time series with hidden markov models. summit.sfu.ca.

[18] Tsay, R. (2005). Analysis of financial time series. books.google.com.

[19] Liao, S., & Chou, S. (2013). Data mining investigation of co-movements on the taiwan and china stock markets for future investment portfolio. Expert Systems with Applications.

[20] Kwak, H., & An, J. (2016). Two tales of the world: Comparison of widely used world news datasets gdelt and eventregistry. Proceedings of the International AAAI Conference on Web and Social Media.

[21] Xia, Y. (2018). Deep learning for financial time series forecasting. dr.ntu.edu.sg.

[22] Schrodt, P., & Yonamine, J. (2012). Automated coding of very large scale political event data. Unable to Determine Complete Publication Venue.

[23] Hoffart, F., D’Orazio, P., Holz, F., & Kemfert, C. (2024). Exploring the interdependence of climate, finance, energy, and geopolitics: A conceptual framework for systemic risks amidst multiple crises. Applied Energy.

[24] Zheng, L., Islam, N., Zhang, J., Behl, A., et al. (2025). Aligning risk and value creation: A process model of supply chain risk management in geopolitical disruptions. International Journal of Operations & Production Management.

[25] Lee, S., & Yoo, S. (2019). Multimodal deep learning for finance: Integrating and forecasting international stock markets. arXiv preprint arXiv:1903.06478.

[26] Li, Y., Zhu, Z., Guo, X., Li, S., Yang, Y., & Zhao, Y. (2023). HGV4Risk: Hierarchical global view-guided sequence representation learning for risk prediction. ACM Transactions on Knowledge Discovery from Data.

[27] Ploeg, F. (1986). Rational expectations, risk and chaos in financial markets. The Economic Journal.