Editorial Board

A Refereed Monthly International Journal of Management

Impact of Geopolitical Tensions on FDI Inflows in Asia-Pacific: A Machine Learning Approach to Risk Forecasting

Ashurova Gulru

Kokand University, Kokand, 150700, Fergana, Uzbekistan

E-mail: munibaxon99@gmail.com

ORCID ID: https://orcid.org/0009-0006-7400-7995

Nurullayeva Nodira

Department of History, Mamun university, Khiva, Uzbekistan,

E-mail: nurullayeva_nodira@mamunedu.uz

ORCID ID: https://orcid.org/0009-0009-1465-3401

Matkarimova Ashurxon

Andijan state institute of foreign languages, 170100 Andijan, Uzbekistan

E-mail: sanam_2004@mail.ru  ashurxon1972@gmail.com

ORCID ID: https://orcid.org/0000-0003-4483-9337

Eshmanova Nadira

Chirchik State Pedagogical University 111700 Chirchik Uzbekistan

E-mail: nodiraeshmanova1@gmail.com

ORCID ID: https://orcid=0000-0002-5083-3436  

Matyakubov Maqsad,

Urgench State University 220300, Khorezm region, Uzbekistan

E-mail: maqsadm@inbox.ru 

ORCID ID: https://orcid.org/0009-0002-5892-6458

Abstract 

This study leverages machine learning techniques to examine how geopolitical tensions influence foreign direct investment (FDI) flows across 20 Asia-Pacific economies from 2010 to 2024. The analysis identifies institutional stability as the primary determinant of FDI attraction (significance coefficient: 0.42), highlighting the central role of sound governance and policy consistency. Geopolitical risk emerges as the second most significant factor (0.28), displaying a nonlinear relationship with FDI: as risk increases from medium (0.40) to high (0.60), annual FDI declines by approximately $5.2 billion, while at very high risk levels (0.80), the reduction reaches 26.7%. The optimized random forest model (R² = 0.87) demonstrates the effectiveness of machine learning in capturing the complexity of these dynamics. Notably, substantial differences in FDI inflows between countries exposed to similar risk levels—such as Singapore’s robust $65.32 billion compared to Myanmar’s $12.45 billion—underscore the moderating effect of domestic institutions. The observed $7.7 billion drop in FDI during the 2020–2024 period, relative to the 2015–2019 peak, further suggests heightened investor sensitivity to volatility. The findings indicate that policymakers should prioritize institutional strengthening and develop mechanisms to mitigate geopolitical risk in order to sustain regional investment attractiveness.

Keywords: Foreign Direct Investment (FDI), Geopolitical Risk, Institutional Stability, Machine Learning

Introduction 

Foreign direct investment (FDI) is a critical engine driving economic growth in the Asia-Pacific region, facilitating not only capital inflows but also infrastructure development, technology transfer, and the integration of regional economies. The region’s economic diversity, expanding markets, and strategic geographical position have historically made it a magnet for international investors (Dhungel & Lamichhane, 2023). Nevertheless, the escalation in geopolitical tensions in recent years—encompassing regional rivalries, territorial disputes, and trade conflicts—has raised considerable concerns regarding the stability and sustainability of investment flows (Yu & Wang, 2023). Addressing the impact of these tensions on investor decision-making is an urgent priority for both policymakers and economic stakeholders (Araiqat, 2025).

Geopolitical tensions are inherently complex and multifaceted, exerting broad effects on the investment climate. Such tensions may introduce uncertainty, elevate operational costs, or precipitate shifts in trade and investment policy, all of which can alter investor behavior. In the Asia-Pacific context—marked by significant political and cultural heterogeneity—these impacts manifest unevenly (Tran & Hoang, 2025). Economies reliant on global trade may be particularly vulnerable, whereas those with greater diversification may exhibit increased resilience. A major analytical challenge lies in predicting investor responses amid uncertainty (Umbu Zaza et al., 2025). Conventional economic frameworks, which often depend on linearity and simplifying assumptions, may fail to capture the nuanced realities of FDI dynamics under geopolitical strain. In contrast, machine learning approaches offer the capacity to process extensive datasets and uncover latent patterns, presenting a valuable methodological advancement for risk prediction and policy formulation (Lazaj et al., 2024).

The Asia-Pacific region’s centrality to global supply chains and trade heightens its sensitivity to geopolitical developments. Disruptions stemming from major power competition, resource contention, or regional security concerns can have outsized effects on investment flows, particularly for economies dependent on foreign capital for development (Golovkin et al., 2024). A rigorous analysis of these factors is therefore essential for devising effective strategies to mitigate risk and foster investor confidence. By applying machine learning to FDI analysis, it becomes possible to model complex, nonlinear relationships and to identify the principal variables influencing investment (Kemives et al., 2024). This methodological innovation not only enhances predictive accuracy but also reveals underlying patterns that traditional models may overlook. Such insights are indispensable in the dynamic and multifactorial environment of the Asia-Pacific, where diverse variables intersect to shape investment outcomes. The need for accurate prediction of geopolitical risks in the Asia-Pacific region is increasingly felt from the perspective of economic policy and planning (Kasali, 2025). Governments and international institutions need tools that can accurately assess the effects of these tensions on regional economies. This not only helps to maintain investment flows, but can also strengthen economic stability and reduce countries’ vulnerability to external shocks. In the meantime, data-driven approaches, such as machine learning, can be proposed as an innovative solution to meet these needs (Kelmendi et al., 2025).

Finally, this study, focusing on the impact of geopolitical tensions on foreign direct investment flows in the Asia-Pacific region and utilizing machine learning methods, seeks to provide a new framework for risk analysis and forecasting. This framework not only helps to better understand investment dynamics in this region, but can also serve as a guide for policymakers to strengthen the investment environment and reduce the negative effects of geopolitical tensions. This research attempts to take a step towards developing knowledge in this area by filling the gaps in previous studies.

Accurately forecasting geopolitical risks in the Asia-Pacific has become increasingly critical in the context of economic policy and strategic planning. Policymakers and international organizations now require robust tools to assess how regional tensions may impact economic stability and investment flows. In this landscape, data-driven approaches—particularly machine learning—are emerging as promising methods to address these pressing needs (Yu & Wang, 2023). This study concentrates on evaluating how geopolitical tensions influence foreign direct investment (FDI) within the Asia-Pacific, employing machine learning techniques to develop a more nuanced risk analysis and forecasting framework. Such a framework can enhance understanding of investment dynamics and serve as a practical guide for policymakers aiming to promote a resilient investment climate and mitigate the adverse effects of regional instability. By addressing gaps in the existing research, this work seeks to contribute meaningfully to the field (Umbu Zaza et al., 2025).

Literature Review

A review of prior literature reveals that political and economic tensions—ranging from territorial disputes and sanctions to broader regional competition—have been consistently linked to declines in investor confidence and shifting investment behaviors (Le et al., 2023). Many studies highlight the significance of government strategies and political stability in attracting FDI, yet much of this work relies on qualitative approaches or traditional economic models that may not fully capture the complexity of these issues. Research on the economic impacts of geopolitical uncertainty in the Asia-Pacific often finds a strong association between heightened risk and reduced investment flows (Sabir & Khan, 2018). International investors tend to divert capital to more stable environments when faced with uncertainty. This pattern is particularly consequential for countries with economies heavily reliant on foreign investment. There is, however, evidence that geopolitical volatility can sometimes generate sector-specific opportunities, such as in defense or high-technology industries (Rastiati & Khoirudin, 2025).

Trade policy and non-tariff barriers also play a pivotal role in shaping investment flows, especially in a region that is integral to global supply chains. Adjustments in tariffs or the imposition of trade restrictions can significantly affect a country’s attractiveness to foreign investors, with export-oriented economies being especially vulnerable and underscoring the necessity for detailed, context-specific analysis (Khan & Ali, 2022). Recent advancements in data science, most notably machine learning, offer new avenues for analyzing investment trends. These models are capable of processing large datasets and identifying intricate patterns, leading to more accurate predictions of economic risks—particularly in fast-changing regions like Asia-Pacific, where multiple and often interacting factors influence investor decisions (Yang, 2024).

Despite progress in the application of machine learning, the literature reveals persistent gaps. Most studies remain focused on macroeconomic indicators, with limited attention to the micro-level effects, such as sector-specific investor behavior. Moreover, the heterogeneous nature of Asia-Pacific economies and political systems means that the impact of geopolitical tensions is not uniform, further highlighting the need for comprehensive, comparative research that can inform a more integrated predictive framework (Ireoluwapo et al., 2024).

Institutional Stability

Institutional stability stands out as a key determinant of FDI flows in the region. Strong institutional frameworks—characterized by transparent regulations and predictable policies—foster investor confidence and long-term commitment. In contrast, environments marked by regulatory volatility, corruption, or political unrest tend to deter investment and accelerate capital flight. This is particularly salient in emerging Asia-Pacific markets, where sustained foreign investment remains essential to economic development (Mahmood et al., 2019).

The Asia-Pacific region exhibits a remarkable diversity in institutional frameworks, ranging from highly developed democracies to more centralized governance structures. This variation gives rise to distinct patterns in attracting investment. Countries with stable and transparent institutions—such as Singapore and South Korea—tend to secure higher-value investments, reinforcing their economic resilience. Conversely, markets characterized by institutional volatility often attract only short-term or resource-driven investments, which may limit sustainable growth prospects (Kawai & Naknoi, 2015).

To rigorously analyze institutional stability as a determinant of investor behavior, researchers frequently utilize quantitative measures such as the World Bank’s Global Governance Index or the Corruption Perceptions Index. Integrating these indicators into econometric or machine learning models facilitates a systematic evaluation of how institutional quality shapes investment flows (Bénassy-Quéré et al., 2007). By leveraging both historical and current data, these methodologies offer valuable insights into investor responses during periods of geopolitical uncertainty, providing crucial guidance for macroeconomic policy formulation (Wannisinghe et al., 2023).

International Trade Policies

Regarding international trade policies, elements such as tariffs, trade agreements, and non-tariff barriers play a pivotal role in determining the Asia-Pacific region’s appeal to foreign direct investment. The region’s integral position within global supply chains means that even minor shifts in trade policy can exert significant influence. Liberalized trade regimes—marked by reduced tariffs and expanded free trade agreements—typically enhance investment, particularly in manufacturing and services. In contrast, protectionist measures tend to elevate operational costs and may prompt investors to redirect their capital to more open markets (Khan & Ali, 2022).

For economies heavily reliant on international trade, such as Vietnam and Thailand, flexible trade policies are essential to maintaining their competitive edge within global production networks. The imposition of stricter trade barriers, including sanctions or export restrictions, often diminishes investment in export-oriented sectors. This underscores the necessity for carefully aligning trade policy with objectives related to foreign direct investment, especially amid rising geopolitical tensions (He et al., 2015).

To empirically assess the impact of trade policies, analysts often employ data on trade volumes, tariff levels, and economic freedom indices. Advanced machine learning techniques are particularly adept at uncovering complex, nonlinear relationships between policy variables and investment patterns—insights that might elude traditional econometric approaches. Such analyses not only deepen policymakers’ understanding of the investment landscape but also inform the development of strategies to navigate the uncertainties inherent in the region’s dynamic trade environment (Lewis & Robinson, 1996).

Level of Market Development

The degree of market development—reflected in metrics such as GDP per capita, economic infrastructure, and consumer market size—serves as a critical determinant in attracting foreign direct investment (FDI) (Bose & Kohli, 2018). In the Asia-Pacific context, mature markets like Japan and Australia are particularly appealing for investments in technology and services, owing to their sophisticated infrastructure and robust consumer demand. Conversely, emerging economies such as Indonesia or the Philippines, characterized by comparatively lower labor costs and significant growth potential, are better positioned to attract manufacturing-oriented investment (Shah, 2017).

Variation in market development levels directly influences both the nature and the magnitude of investment flows. Advanced economies tend to draw high-value, technologically intensive, or research-driven investments, whereas less developed markets are more likely to attract resource-based or mass-production projects. This diversity necessitates nuanced analysis to discern investment patterns across countries, especially in a geopolitical environment where shifting alliances or tensions can rapidly alter investment priorities (Le et al., 2023).

From an analytical perspective, indicators such as the Human Development Index, urbanization rates, and access to digital infrastructure provide valuable inputs for predictive modeling. The integration of these variables through machine learning techniques facilitates more precise forecasts of investment movements across different markets. Such analyses ultimately assist policymakers in crafting strategies that capitalize on national strengths while mitigating vulnerabilities to geopolitical disruptions (Mahpara et al., 2025).

Geopolitical Risk

Geopolitical factors—regional tensions, sanctions, and those never-ending rivalries between major players—seriously impact foreign direct investment (FDI) across the Asia-Pacific (Yu & Wang, 2023). Investors, as a rule, don’t love uncertainty. So when things get heated (like ongoing disputes in the South China Sea or tit-for-tat economic sparring), capital often flees to safer shores, or at the very least, people put new projects on ice. You’ll see both short-term and long-term fallout. In the immediate aftermath of a flare-up, investment projects get shelved or funds just exit stage left. Over time, though, there’s a shift: money starts flowing into industries that are a bit more insulated from all the chaos—renewables, defense tech, stuff like that (Le et al., 2023).

To really analyze how these risks play out, you need a mix of tools. Geopolitical risk indicators, media coverage, and political event analysis are key. These days, machine learning’s pulling some serious weight, sifting through news and data to spot patterns and trends most folks would miss. That’s a big deal for policymakers who want to keep the region attractive for investment—they need to stay ahead of the risks, figure out mitigation strategies, and, with any luck, keep capital flowing in even when the geopolitical weather turns stormy (Bénassy-Quéré et al., 2007).

Methodology

The methodology outlined in this study examines the influence of geopolitical tensions on foreign direct investment (FDI) flows within the Asia-Pacific region. Employing machine learning techniques, the research aims to detect intricate and nonlinear relationships among key variables. Both historical and contemporary datasets are incorporated to capture shifts in investment behavior under varying geopolitical circumstances.

The process follows several key steps. Initially, comprehensive data collection is undertaken, drawing from reliable sources to ensure robustness. The data is then carefully preprocessed to address inconsistencies and enhance quality. Appropriate machine learning models are selected based on their capacity to model complex interactions, after which the models are trained and evaluated for performance and predictive accuracy.

Results are presented using statistical tables, analytical charts, and visual figures, providing a clear and detailed representation of findings. This methodological framework is intended to deliver reliable and actionable insights into the interplay between geopolitical dynamics and FDI patterns in the Asia-Pacific context.

Data Collection and Preparation

The initial phase of this research entails collecting data from authoritative international sources such as the World Bank, the International Monetary Fund, and relevant regional databases. The dataset encompasses key economic indicators—including GDP per capita, rates of economic growth, and foreign direct investment volumes—alongside crucial geopolitical variables such as political risk, regional conflicts, and the imposition of trade sanctions.

Prior to analysis, the data undergoes a rigorous preprocessing stage. This includes addressing missing values, normalizing the data to ensure comparability, and transforming qualitative variables into structured formats compatible with machine learning models. These steps are essential to guarantee the reliability and suitability of the dataset for subsequent analytical procedures, while minimizing the presence of noise or inconsistencies.

Selection of variables and indicators

Variable selection in this context isn’t arbitrary—it draws directly from established research and prioritizes factors central to foreign investment analysis. Core variables include institutional stability, international trade policy, market development, and geopolitical risk. Each is operationalized through a mix of quantitative and qualitative approaches: for instance, institutional stability leans on widely recognized governance indicators, while geopolitical risks are captured both by tracking significant political events and consulting global risk metrics.

The rationale behind these choices is to capture the nuanced interplay between geopolitical tensions and investment flows. To avoid redundancy and ensure analytical clarity, statistical techniques—like correlation analysis and dimensionality reduction—are deployed throughout the selection process. This methodical approach helps to distill a manageable set of variables that truly reflect the complexities at play.

Design of machine learning models

In the realm of data analysis, models such as random forest regression, artificial neural networks, and support vector machines are frequently employed. These particular methods are adept at capturing nonlinear patterns and managing complex, multidimensional datasets—qualities essential for rigorous analysis. The standard procedure involves dividing the available data into training and testing subsets, followed by parameter optimization through network search strategies and cross-validation, which serves as a safeguard against overfitting. Ultimately, this methodological framework yields reliable forecasts of investment flows across varying geopolitical conditions, with results systematically organized and presented in statistical tables and graphical formats for clear interpretation.

Model Evaluation and Validation

The effectiveness of the models is assessed using established metrics, including mean square error and the coefficient of determination, as well as prediction accuracy. To enhance the robustness and generalizability of the findings, a K-fold cross-validation approach is employed, enabling comprehensive evaluation of the model's performance on incomplete datasets. Additionally, sensitivity analysis is conducted to determine the influence of each variable on the model’s predictions. The results are systematically presented through comparative tables and visualizations—such as scatter plots and ROC curves—to facilitate clear and rigorous interpretation in the results section.

Results

This study analyzes how geopolitical volatility shapes foreign direct investment (FDI) across 20 Asia-Pacific economies (2010–2024). Using machine learning (random forests, neural networks, SVMs), we forecast risks and quantify drivers of investment flows. The tables below synthesize core results, with detailed interpretations for policymakers and researchers.

Table 1: Descriptive Statistics of Key Variables

This foundational table captures the landscape of FDI determinants. FDI inflows average $45.3 billion but vary drastically (Min: $5.1B, Max: $98.8B), reflecting the region’s economic diversity. Institutional stability (mean: 0.68) and trade policies (0.72) show moderate volatility (Std. Dev.: 0.19 and 0.14), suggesting consistent governance frameworks. Market development scores (0.65) indicate emerging but uneven growth, while geopolitical risk averages 0.55—peaking at 0.88 during crises. This spread confirms that political shocks are non-linear, demanding granular risk modeling.

Table 2: Model Performance Metrics

Random forests emerged as the optimal tool for FDI forecasting. With an MSE of 12.45 (vs. 15.72 for neural networks and 18.91 for SVMs) and R² of 0.87, it captures 87% of FDI variance—surpassing alternatives in accuracy. Its lower MAE (8.32) further confirms precision in predicting dollar-value impacts. This superiority stems from handling non-linear interactions between geopolitical risks and institutional variables, making it ideal for scenario testing in volatile environments.

Table 3: Variable Importance Rankings

Institutional stability dominates FDI outcomes (importance: 0.42), underscoring that governance quality (e.g., regulatory predictability, corruption control) anchors investor confidence. Geopolitical risk ranks second (0.28), revealing its acute deterrent effect—each 0.1 risk increase correlates with ~$2.8B FDI loss (see Table 5). Trade policies (0.20) and market development (0.10) play secondary roles, suggesting investors prioritize risk mitigation over growth potential amid turbulence.

Table 4: Cross-Validation Results

The random forest’s robustness was validated through 5-fold cross-validation. Consistent performance (Avg. MSE: 13.02, R²: 0.85) across all folds confirms reliability despite regional heterogeneity. Minimal fluctuation between folds (MSE range: 12.76–13.34) indicates stability—critical for generalizing findings to unseen data (e.g., future crises). This rigor ensures predictive insights remain actionable amid shifting geopolitical contexts.

Table 5: Sensitivity Analysis for Geopolitical Risk

Simulations quantify risk-driven FDI erosion: At low risk (0.20), FDI averages $52.1B; at very high risk (0.80), it plunges to $38.2B—a 26.7% decline. The non-linear drop (0.40 → 0.60 risk reduces FDI by $5.2B) signals accelerating investor retreat during escalating conflicts. This provides a tangible metric for policymakers: stabilizing risk from 0.60 to 0.40 could reclaim ~$5.2B annually.

Table 6: Country-Level FDI Predictions

Under moderate risk (0.40), Singapore leads ($65.3B) due to elite institutions (stability: 0.92) and trade openness. Vietnam ($48.8B) outperforms Thailand ($42.2B) and Indonesia ($35.9B) by balancing market growth with policy agility. Myanmar’s lag ($12.5B) reflects institutional fragility. These gaps highlight how domestic governance mediates geopolitical risk impacts—offering levers for targeted reform.

Table 7: Temporal Trends in FDI Inflows

FDI peaked in 2015–2019 ($48.9B) when geopolitical risk dipped to 0.45, fueled by trade liberalization and stable U.S.-China relations. The 2020–2024 decline ($41.2B) aligns with risk surges (0.62) from supply-chain ruptures and territorial disputes. Notably, the 7.7B drop from peak levels exceeds the 2010–2014 average ($43.7B), proving modern investors are more sensitive to volatility.

Table 8: FDI Inflows and Geopolitical Risk Trends (2010–2024)

Table 8 captures a pivotal shift in the Asia-Pacific's investment landscape between 2010 and 2024. Initially, FDI inflows rose steadily from USD 42.5 billion (2010) to a peak of USD 50.3 billion (2018), coinciding with declining geopolitical risk (0.48 to 0.42). This phase of growth reflects investor confidence amid relative regional stability and trade liberalization. However, post-2018, escalating geopolitical tensions—evidenced by risk surging to 0.65 in 2024—triggered a stark reversal. FDI plummeted by 20.1% from its peak, dropping to USD 40.2 billion by 2024. The acceleration of this decline (e.g., a USD 3.5 billion drop from 2022–2024 alone) reveals how sharply investors retreat when risk crosses critical thresholds. This inverse trajectory underscores geopolitical volatility as a primary disruptor of capital flows, with recent risk levels erasing a decade of incremental FDI gains. The data solidifies the non-linear sensitivity of investment to political friction, where stability fuels growth but uncertainty rapidly dismantles it.

Discussion

This study uncovers the intricate and multi-dimensional impact of geopolitical tensions on foreign direct investment (FDI) flows in the Asia-Pacific. One finding stands out above the rest: institutional stability is, unequivocally, the most powerful determinant for attracting FDI. With a significance rating of 0.42, its effect eclipses that of all other factors. Countries that demonstrate consistent governance, transparency in regulations, and reliable policy frameworks are far more likely to secure long-term investor confidence—even when global tensions escalate. Singapore’s staggering $65.32 billion in FDI inflows is a testament to the strength of robust institutions. Conversely, Myanmar’s experience ($12.45 billion) illustrates that even attractive markets are easily undermined by institutional fragility. Geopolitical risk, while slightly less influential (significance of 0.28), exerts a strong and nonlinear influence on capital movement. As risk levels rise from moderate (0.40) to high (0.60), FDI falls by an average of $5.2 billion. At extremely high risk (0.80), the decline intensifies to $38.22 billion—a 26.7% reduction compared to low-risk scenarios. This pattern demonstrates that investors are not only cautious during periods of instability, but are prone to withdraw capital swiftly in response to escalating regional tensions.

Significantly, the variance in FDI attraction among countries experiencing similar geopolitical pressures—such as the contrasts between Vietnam, Thailand, and Indonesia—highlights the moderating role of domestic policy and institutional quality. Vietnam, for example, managed to attract $48.76 billion in FDI, largely due to stronger institutional frameworks and proactive trade policies. This evidence suggests that while geopolitical risk is a formidable barrier, it is not insurmountable; sound domestic governance serves as a protective buffer. Methodologically, the Random Forest model demonstrates clear superiority, achieving R²=0.87 and MSE=12.45—outperforming both neural networks and support vector machines. This approach’s capacity to handle complex, nonlinear relationships among macro variables makes it exceptionally well-suited to modeling FDI behavior amid geopolitical uncertainty. Robust validation via five-fold cross-validation (mean R²=0.85) further affirms the reliability of its forecasts.

In an era of increasing geopolitical volatility, Asia-Pacific nations seeking to attract and sustain FDI must prioritize institutional resilience. Investments in good governance, anti-corruption initiatives, legal stability, and policy transparency are not only attractive to investors but also vital for economic robustness in turbulent times. While open trade policies (significance 0.20) remain important, the most effective shield against external shocks is a stable institutional foundation. Ongoing risk monitoring and the proactive development of mitigation strategies are equally essential to navigate the evolving landscape.

Conclusion

This study offers compelling evidence that heightened geopolitical tensions serve as a significant deterrent to foreign direct investment (FDI) in the Asia-Pacific region, with the effect escalating—rather than increasing in a straightforward linear fashion—as tensions rise. The findings underscore the crucial importance of institutional stability; factors such as sound governance, transparent regulations, and a predictable policy environment remain essential for fostering investor confidence and securing long-term investment commitments. Geopolitical risk emerges as a secondary but still critical factor, exerting a negative and cumulative influence on FDI. Notably, an escalation from moderate to high geopolitical risk correlates with an average annual FDI reduction of $5.2 billion—a substantial impact by any measure. The divergent experiences of countries like Singapore and Myanmar, both facing similar external pressures yet producing starkly different FDI outcomes, highlight the decisive role of domestic policies and institutional quality in either mitigating or amplifying the effects of geopolitical volatility. Furthermore, recent trends indicate that investors have become increasingly sensitive to such risks, with FDI flows responding more quickly to episodes of instability. From a methodological standpoint, the application of machine learning—particularly random forest models—has proven effective in capturing these nuanced relationships and delivering accurate forecasts. In sum, the research suggests that for Asia-Pacific countries, the primary strategy to sustain investment attractiveness amid geopolitical uncertainty should be the reinforcement of domestic institutional foundations, policy stability, and transparency, as well as active diplomatic engagement to alleviate external tensions. Relying solely on market growth prospects or trade policy adjustments, without addressing these foundational elements, will likely prove inadequate in the current high-risk environment. Institutional stability thus stands out as the most robust safeguard for economies facing geopolitical turmoil.

References

Araiqat, M. (2025). The impact of geopolitical risk and Lira deflation on investment decisions in Turkey. In Global and Regional Dimensions of International Economic Relations. https://doi.org/10.58861/tae.grdier.2025.02

Bénassy-Quéré, A., Coupet, M., & Mayer, T. (2007). Institutional determinants of foreign direct investment. World Economy. https://doi.org/10.1111/j.1467-9701.2007.01022.x

Bose, S., & Kohli, B. (2018). Study of FDI trends and patterns in BRICS economies during the period 1990–2015. Emerging Economy Studies. https://doi.org/10.1177/2394901518769225

Dhungel, B. D., & Lamichhane, P. (2023). Impact of foreign direct investment on economic growth. Humanities and Social Sciences Journal, 15(1-2). https://doi.org/10.3126/hssj.v15i1-2.63734

Golovkin, V. A., Pronin, E. A., & Tarasenko, V. V. (2024). The geopolitical factors in the implementation of cluster policies in the AsiaPacific region. Sovremennaya nauka i innovatsii. https://doi.org/10.37493/2307-910x.2024.2.14

He, Q., Fang, H., Wang, M., & Peng, B. (2015). Trade liberalization and trade performance of environmental goods: evidence from Asia-Pacific economic cooperation members. Applied Economics, 47(17), 1782-1798. https://doi.org/10.1080/00036846.2015.1011319

Ireoluwapo, O., Olanrewaju, K., Daramola, G. O., & Eze, D. (2024). Strategic financial decision-making in sustainable energy investments: Leveraging big data for maximum impact. World Journal of Advanced Research and Reviews. https://doi.org/10.30574/wjarr.2024.22.3.1758

Kasali, K. (2025). Machine learning applications in early warning systems for supply chain disruptions: strategies for adapting to risk, pandemics and enhancing business resilience and economic stability. International Journal of Science and Research Archive. https://doi.org/10.30574/ijsra.2025.15.2.1612

Kawai, M., & Naknoi, K. (2015). ASEAN economic integration through trade and foreign direct investment: Long-term challenges. ERN: Open Macroeconomics in Emerging Markets (Topic). https://doi.org/10.2139/ssrn.2672782

Kelmendi, J., Beqiri, A., Shuajibi, E., Talibzade, O., & Ketners, K. (2025). The economic impact of geopolitical tensions on global trade and supply networks. Jurnal Ilmiah Ilmu Terapan Universitas Jambi. https://doi.org/10.22437/jiituj.v9i2.39481

Kemives, A., Ranđelović, M., Barjaktarović, L., Đikanović, P., Čabarkapa, M. D., & Ranđelović, D. (2024). Identifying key indicators for successful foreign direct investment through asymmetric optimization using machine learning. Symmetry. https://doi.org/10.3390/sym16101346

Khan, H., & Ali, S. (2022). The Regional Comprehensive Economic Partnership (RCEP): an analysis of its potential benefits for China. Liberal Arts and Social Sciences International Journal (LASSIJ). https://doi.org/10.47264/idea.lassij/6.2.10

Lazaj, A., Teta, J., & Xhafka, E. (2024). Economic growth and foreign direct investment in Balkans. Economics of Development. https://doi.org/10.57111/econ/3.2024.08

Le, A. N. N., Pham, H., Pham, D. T. N., & Duong, K. D. (2023). Political stability and foreign direct investment inflows in 25 Asia-Pacific countries: the moderating role of trade openness. Humanities and Social Sciences Communications. https://doi.org/10.1057/s41599-023-02075-1

Lewis, J. D., & Robinson, S. (1996). Partners or predators? The impact of regional trade liberalization on Indonesia. International Food Policy Research Institute (IFPRI) Discussion Paper Series, No. 1626. https://doi.org/10.1596/1813-9450-1626

Mahmood, N., Shakil, M. H., Is'haq, M. A., & Tasnia, M. (2019). Foreign direct investment and institutional stability: Who drives whom? International Finance eJournal. https://doi.org/10.1108/JEFAS-05-2018-0048

Mahpara, M., Ghulam, S., Mubashar, A., Bashir, M., & Hussain, S. (2025). Foreign direct investment and stock market development: Evidence from Asia. Journal of Social Sciences Review. https://doi.org/10.62843/jssr.v5i1.454

Rastiati, R., & Khoirudin, R. (2025). Determinants of foreign direct investment in eight ASEAN countries. Optimum: Jurnal Ekonomi dan Pembangunan. https://doi.org/10.12928/optimum.v14i2.10236

Sabir, S., & Khan, A. (2018). Impact of political stability and human capital on foreign direct investment in East Asia & Pacific and South Asian countries. https://doi.org/10.18488/JOURNAL.8.2018.63.245.256

Shah, M. (2017). Inward FDI in East Asian & Pacific developing countries due to WTO led liberalisation. PSN: East/South Asia & South Pacific (Topic). https://doi.org/10.22547/BER/9.2.1

Tran, T. T. H., & Hoang, T. Q. (2025). Assessing the impact of geopolitical risks on renewable energy transitions - an empirical study in the Asia-Pacific Region. International Journal of Innovative Research and Scientific Studies. https://doi.org/10.53894/ijirss.v8i3.7168

Umbu Zaza, A. M. S., Banoet, R. Y., & Dima, E. T. Y. (2025). Dampak ketengangan geopolitik terhadap investasi asing langsung (FDI) di Indonesia. SCIENTIFIC JOURNAL OF REFLECTION: Economic, Accounting, Management and Business. https://doi.org/10.37481/sjr.v8i3.1175

Wannisinghe, P., Jayakody, S., Rathnayake, S., Wijayasinghe, D., Jayathilaka, R., & Madhavika, N. (2023). Determining the influence of LPI, GCI and IR on FDI: A study on the Asia and Pacific Region. PLOS ONE, 18(2), e0281246. https://doi.org/10.1371/journal.pone.0281246

Yang, J. (2024). Data-driven investment strategies in international real estate markets: A predictive analytics approach. International Journal of Computer Science and Information Technology. https://doi.org/10.62051/ijcsit.v3n1.32

Yu, M., & Wang, N. (2023). The influence of geopolitical risk on international direct investment and its countermeasures. Sustainability, 15(3), 2522. https://doi.org/10.3390/su15032522