تاثیر ریسک های ژئوپلیتیکی بر بازده ارزهای دیجیتال با استفاده از الگوریتم های بهینه سازی
محورهای موضوعی : فصلنامه اقتصاد محاسباتی
ساناز عباسی
1
,
زهرا هنرمندی
2
,
زهرا فرشادفر
3
1 - دانشجو
2 - گروه حسابداری.واحد تهران غرب. دانشگاه آزاد اسلامی. تهران.ایران
3 - عضو هیات علمی
کلید واژه: ریسک های ژئوپلیتیکی, بازده ارزهای دیجیتال , الگوریتم بهینه سازی علف های هرز, الگوریتم بهینه سازی توده ذرات,
چکیده مقاله :
ارزیابی تأثیر ریسکهای ژئوپلیتیکی بر بازده ارزهای دیجیتال بهدلیل ماهیت غیرخطی، پویا و چندبُعدی این رابطه، نیازمند روشهای تحلیلی پیشرفته است. الگوریتمهای بهینهسازی هوشمند نظیر الگوریتم علفهای هرز و الگوریتم توده ذرات بهعنوان ابزارهای کارآمد در این حوزه عمل میکنند. آنها میتوانند با جستجوی کارآمد در فضای پارامترهای مؤثر، بهترین ترکیب شاخصهای ژئوپلیتیکی و تأثیر آنها بر نوسانات بیتکوین را شناسایی کنند، بدون آنکه به مدلهای خطی محدود باشند.
بنابراین در این تحقیق به ارزیابی تاثیر ریسک های ژئوپلیتیکی بر بازده ارزهای دیجیتال با استفاده از الگوریتم های بهینه سازی علف های هرز و توده ذرات پرداخته شد. این پژوهش، برای دوره زمانی از 01/01/1402 الی 31/04/1404 به صورت روزانه صورت گرفت. روش آماری مورد استفاده در این تحقیق روش رگرسیون چند متغیره به شیوه پانل دیتا و الگوریتم بهینه سازی شده می باشد. نتایج حاصل شده از تحقیق نشان داد که 1- بین ریسک های ژئوپلیتیکی با بازدهی بیت کوین رابطه معنی داری به صورت معکوس وجود دارد.، 2- الگوریتم بهینه سازی علف های هرز توان پیش بینی بازدهی بیت کوین را با استفاده از مولفه های ریسک های ژئوپلیتیکی دارد.، 3- الگوریتم بهینه سازی توده ذرات توان پیش بینی بازدهی بیت کوین را با استفاده از مولفه های ریسک های ژئوپلیتیکی دارد.، 4- الگوریتم بهینه سازی علف های هرز نسبت به الگوریتم بهینه سازی توده ذرات جهت پیش بینی بازدهی بیت کوین با استفاده از مولفه های ریسک های ژئوپلیتیکی توان بالاتری برای پیش بینی را دارا می باشد.
Extended Abstract
Purpose
The contemporary global landscape is characterized by accelerated and multifaceted geopolitical shifts, where the traditional boundaries between economics, politics, and technology have become profoundly intertwined and increasingly indistinct. This complex interconnectivity creates a volatile environment where macroeconomic policies, international sanctions, and regional instabilities generate cascading effects across all asset classes. Within this turbulent milieu, cryptocurrencies have rapidly evolved from an obscure digital experiment into a novel and significant phenomenon within the global financial system, challenging established monetary paradigms and institutional structures. Initially heralded by proponents as a digital "safe haven" asset class, this perception was rooted in their core architectural principles: decentralization, cryptographic transparency, censorship resistance, and operational independence from traditional banking and governmental systems.However, as the asset class has matured and its market capitalization has grown, its susceptibility to pronounced volatility stemming from exogenous geopolitical tensions has become increasingly evident and a critical subject of scholarly inquiry. Events such as escalating trade wars, the imposition of international sanctions, acute regional crises, and sudden, unpredictable shifts in government macro-policies and regulatory stances have consistently demonstrated their capacity to induce significant price dislocations in the cryptocurrency markets. This apparent paradox—a decentralized asset reacting strongly to centralized power dynamics—necessitates a move beyond simplistic, linear analytical paradigms.Consequently, this study seeks to address this complexity by employing advanced econometric methodologies to investigate the intricate, non-linear, and dynamic causal relationships between quantifiable geopolitical risks (GPR) and the returns of major cryptocurrencies, with a primary empirical focus on Bitcoin (BTC). The central research objective is to rigorously determine and empirically validate whether cryptocurrencies, particularly Bitcoin, can effectively act as a strategic risk-hedging asset or a reliable store of value during periods of acute geopolitical crisis, or whether, conversely, they ultimately become victims of the very volatility and risk-aversion they were designed to circumvent, thus behaving more like speculative risk-on assets within the broader financial ecosystem.
Methodology
This applied research employs a descriptive, positivist methodology with deductive-inductive reasoning, utilizing ex-post facto (past event) data. The study period spans from 01/01/2023 to 07/22/2025, using daily data. The statistical population comprises two groups: (1) cryptocurrency data (Bitcoin) and (2) the Global Geopolitical Risk (GPR) index, sourced daily from the established database at https://www.matteoiacoviello.com/gpr.htm.
A multivariate panel data regression model was first used to test the baseline relationship:
BTCit=β0+β1GPR i,t+β2Oil i,t+β3Inflation it+β4GOLDit+εitBTCit =β0 +β1 GPR i,t +β2 Oil i,t +β3 Inflation it +β4 GOLDit +εit
Where the dependent variable is Bitcoin returns, the independent variable is the natural logarithm of the daily GPR, and control variables include changes in oil prices, global inflation rate, and changes in global gold prices. Subsequently, to enhance predictive accuracy and model the non-linear dynamics, two metaheuristic optimization algorithms were implemented in MATLAB: the Invasive Weed Optimization (IWO) algorithm and the Particle Swarm Optimization (PSO) algorithm. The input variables for these AI-driven predictive models were the components of geopolitical risk.
Finding
The key findings of the study are as follows:
- Hypothesis 1: A significant inverse relationship was found between geopolitical risks and Bitcoin returns (β = -0.245, p-value = 0.000). This suggests that during periods of high geopolitical instability, Bitcoin's price and returns often increase, positioning it as a potential "digital safe-haven" asset.
- Hypothesis 2 & 3: Both the IWO and PSO algorithms demonstrated a high capability to predict Bitcoin returns using geopolitical risk components, confirming their utility as advanced analytical tools for this complex relationship.
- Hypothesis 4: A comparative analysis revealed the superior predictive performance of the IWO algorithm over the PSO algorithm. The IWO algorithm achieved a higher predictive accuracy (98.07%) with a lower mean squared error (MSE = 1.444), compared to the PSO algorithm (93.78% accuracy, MSE = 1.840). The IWO's adaptive mechanisms, including non-uniform seed dispersal and competitive selection, proved more effective in navigating the noisy, non-linear data space of geopolitical risks.
Pre-estimation tests confirmed the stationarity of all variables (via Fisher test), the absence of multicollinearity (VIF values < 5 for all variables), and homoscedasticity of residuals (Breusch-Pagan test, p-value = 0.114).
Conclusion
This research makes a significant contribution by demonstrating that Bitcoin exhibits a significant and immediate inverse reaction to geopolitical shocks, a finding that contrasts with some previous studies using linear models that found no long-term relationship. The results align with the narrative of Bitcoin being a decentralized, censorship-resistant asset with a capped supply, making it attractive during times of traditional market uncertainty. Methodologically, the study introduces a novel application of bio-inspired optimization algorithms, particularly IWO, in the field of financial econometrics and cryptocurrency analysis. The superior performance of IWO is attributed to its robust exploration and exploitation capabilities in complex, high-dimensional problem spaces typical of geopolitical and financial data.
Practical Implications and Future Research
- Investors: Can consider allocating a small percentage (1-5%) of their portfolio to Bitcoin as a potential hedge against geopolitical turmoil, while monitoring confounding variables like government regulations.
- Policymakers: In crisis-prone countries, understanding this relationship is crucial for market stability and could inform policies around digital assets as a means to mitigate the impact of international sanctions.
- Future Research: Should focus on integrating the IWO algorithm with deep learning techniques (e.g., LSTM networks), expanding the analysis to other cryptocurrencies and asset classes, and developing more standardized, high-frequency geopolitical risk datasets.
References
Aksoy-Hazır, Ç. And Tan, O. F. (2023). The impact of geopolitical risk on cash holdings policy: evidence from an emerging market. Managerial Finance, 49(9), 1400-1419. Retrieved from https://doi.org/10.1108/MF-01-2022-0024/
Alam, A. W., Houston, R. and Farjana, A. (2023). Geopolitical risk and corporate investment: How do politically connected firms respond?. Finance Research Letters, 53, (103681),1-16. Retrieved from https://doi.org/10.1016/j.frl.2023.103681/
Buthelezi, E. M. (2024). Navigating global uncertainty: Examining the effect of geopolitical risks on cryptocurrency prices and volatility in a Markov-switching vector autoregressive model. International Economic Journal, 38(4), 564-590. Retrieved from https://doi.org/10.1080/10168737.2024.2393589/
Caldara, D. and Iacoviello, M. (2022). Measuring geopolitical risk. American Economic Review, 112(4), 1194-1225. Retrieved from https://doi:10.1257/aer.20191823/
Chen, Y., Chiu, J., Chung, H. and Lien, D. (2024). Bitcoin market connectedness across political uncertainty. International Review of Economics & Finance, 96, 1-26. Retrieved from https://doi.org/10.1016/j.iref.2024.103623/
Cho, J. H. (2024). The effect of geopolitical risk on corporate cash holdings: evidence from Korea. Applied Economics Letters, 31(13), 1159-1165. Retrieved from https://doi.org/10.1080/13504851.2023.2176442/
Darini, V. M. and Namdar Jouimi, E. Study of Global Geopolitical Risks in the Policies and Economic Management of the Islamic Republic of Iran. Quarterly Journal of Political Research in Islamic World, 8 (2), 81-93. Retrieved from https://doi:10.21859/priw-080204/ (In Persian).
Demir, E., Gozgor, G. and Paramati, S. R. (2019). Do geopolitical risks matter for inbound tourism?. Eurasian Business Review, 9, 183-191. Retrieved from https://doi.org/10.1007/s40821-019-00118-9/
Fang, Y., Tang, Q. and Wang, Y. (2024). Geopolitical risk and cryptocurrency market volatility. Emerging Markets Finance and Trade, 60(14), 3254-3270. Retrieved from https://doi.org/10.1080/1540496X.2024.2343948/
Fereydooni, A. and Hajizadeh, E. (2024). Safe haven opportunities for cryptocurrencies in geopolitically risky environments. Applied Economics, 56(58), 8378-8393. Retrieved from https://doi.org/10.1080/00036846.2023.2290592/
Golkhandan, A. and Mohammadian Mansour, S. (2023). Investigating the Symmetric and Asymmetric Causality between Global Geopolitical Risk, Crude Oil Price and Middle East Region's Defense Expenditures. Scientific Journal Quarterly of Middle East Studies, 30(2), 129-161. Retrieved from https://DOR:20.1001.1.15601986.1402.30.2.6.1/ (In Persian).
Goodell, J. W., I. Alon, L. Chiaramonte, A. Dreassi, A. Paltrinieri, and S. Pis era (2023). Risk substitution in cryptocurrencies: evidence from BRICS announce ments, Emerging Markets Review, 54, 1-24. Retrieved from https://doi.org/10.1016/j.ememar.2022.100938/
Gupta, G. (2023). Geopolitical risk and investment-cash flow sensitivity: An empirical analysis for Indian business group-affiliated firms and non-business group-affiliated firms. Finance Research Letters, 58, (104574), 1-12. Retrieved from https://doi.org/10.1016/j.frl.2023.104574/
Kamal, M. R. and Wahlstrøm, R. R. (2023). Cryptocurrencies and the threat versus the act event of geopolitical risk. Finance Research Letters, 57, 1-8. Retrieved from https://doi.org/10.1016/j.frl.2023.104224/
Khalfaoui, R., Gozgor, G. and Goodell, J. W. (2023): Impact of Russia-Ukraine war attention on cryptocurrency: Evidence from quantile dependence analysis, Finance Research Letters, 52, 1-29. Retrieved from https://doi.org/10.1016/j.frl.2022.103365/
Kotcharin, S. and Maneenop, S. (2020). Geopolitical risk and corporate cash holdings in the shipping industry. Transportation Research Part E: Logistics and Transportation Review, 136, 1-16. Retrieved from https://doi.org/10.1016/j.tre.2020.101862/
Lee, C. C., Wang, C. W., Thinh, B. T. and Purnama, M. Y. I. (2023). Cash holdings and cash flows: Do oil price uncertainty and geopolitical risk matter?. Economic Analysis and Policy, 79, 134-152. Retrieved from https://doi.org/10.1016/j.eap.2023.06.013/
Lee, C. C. and Wang, C. W. (2021). Firms' cash reserve, financial constraint, and geopolitical risk. Pacific-Basin Finance Journal, 65, 1-24. Retrieved from https://doi.org/10.1016/j.pacfin.2020.101480/
Long, H., Demir, E., B æEdowska-SÛjka, B. Zaremba, A. and Shahzad, S. J. H. (2022). Is geopolitical risk priced in the cross-section of cryptocurrency returns?, Finance Research Letters, 49, 1-8. Retrieved from https://doi.org/10.1016/j.frl.2022.103131/
Sharifmoghaddasi, A. and Mohammadloo, E. (2025). Investigating the Impact of Geopolitical Risk and the COVID-19 Pandemic on Bitcoin Yields and Volatility.Journal of Program and Development Research, 6(1), 97-122. Retrieved from https://doi.org/10.22034/pbr.2025.465644.1442/ (In Persian).
Umar, Z., Bossman, A., Choi, S.-Y. and Teplova, T. (2022). Does geopolitical risk matter for global asset returns? Evidence from quantile-on-quantile regression. Finance Research Letters, 48, 1-25. Retrieved from https://doi.org/10.1016/j.frl.2022.102991/
Wang, X., Wu, Y. and Xu, W. (2024). Geopolitical risk and investment. Journal of Money, Credit and Banking, 56(8), 2023-2059. Retrieved from https://doi.org/10.1111/jmcb.13110/
Wang, K. H., Xiong, D. P., Mirza, N., Shao, X. F. and Yue, X. G. (2021). Does geopolitical risk uncertainty strengthen or depress cash holdings of oil enterprises? Evidence from China. Pacific-Basin Finance Journal, 66, 1-23. Retrieved from https://doi.org/10.1016/j.pacfin.2021.101516/
Yilmazkuday, H. (2025). Geopolitical risks and cryptocurrency returns. Review of Financial Economics, 43(2), 166-191. Retrieved from https://doi.org/10.1002/rfe.1223/
