IDEAS home Printed from https://ideas.repec.org/a/alk/jefas0/v7y2025i5id10822.html

AI-Powered Early Warning Systems for Emerging Market Crises: Enhancing U.S. Foreign Investment Risk Strategy

Author

Listed:
  • Aktaruzzaman Kafi

    (Department of Computer Science, City University of New York (CUNY), New York, USA)

  • S M Shamsil Arefin

    (Bachelor of Science in Computer Science, CUNY Brooklyn College, USA)

  • Nigar Sultana

    (Department of Finance and Financial Analytics, University of New Haven CT, USA)

  • Md Sanjit Hossain

    (Master of Business Administration, University of the Potomac, Washington DC, USA)

  • Mainuddin Adel Rafi

    (Master of Science/Information System, Pacific State University, USA)

  • Mohammad Mahmudur Rahman

    (Master of Science in Computer Science, Pacific State University, USA)

  • Rafi Muhammad Zakaria

    (Department of Management Science and Information Systems, University of Massachusetts Boston, Boston, USA)

Abstract

Emerging markets (EMs) exhibit nonlinear dynamics and contagion pathways that can rapidly amplify local shocks into systemic crises, exposing U.S. investors to outsized downside risk. This paper proposes an AI-powered early warning system (EWS) that forecasts the probability and timing of EM crisis events currency crashes, sovereign distress, and capital-flow sudden stops over 3/6/12-month horizons. The framework integrates multi-modal data: macro–financial indicators (FX, rates, CDS, reserves, external balances), market microstructure signals (order-flow imbalance, jump intensity), and alternative data (news and social sentiment, trade/shipping activity, satellite night-lights). Methodologically, we combine temporal transformers for regime-aware sequence modeling with graph neural networks over exposure networks (trade, banking, portfolio flows) to capture spillovers, and gradient-boosting models for calibrated probabilities. A composite Crisis Risk Index (CRI) is produced via isotonic calibration, with uncertainty bands from conformal prediction. Model transparency is ensured through SHAP-based global and local explanations, counterfactual analysis for policy levers (e.g., reserve adequacy, rates), and stability checks against data revisions. Backtests benchmark against canonical EWS rules and logistic/KLR-style baselines, evaluating AUC, Brier score, precision, false-alarm costs, and average lead-time. We illustrate decision utility for U.S. foreign investment strategy through three use cases: (i) dynamic country allocation and hedging, (ii) pre-trade risk budgeting with crisis-conditioned scenarios from a generative stress engine, and (iii) portfolio-level loss mitigation under liquidity and currency constraints. The system operationalizes an end-to-end pipeline ingestion, nowcasting, horizon forecasting, and alert governance suitable for investment committees and risk offices. Results indicate materially improved early warning lead-time and fewer false positives versus traditional indicators, enabling earlier de-risking and more resilient U.S. exposure to EM cycles.

Suggested Citation

  • Aktaruzzaman Kafi & S M Shamsil Arefin & Nigar Sultana & Md Sanjit Hossain & Mainuddin Adel Rafi & Mohammad Mahmudur Rahman & Rafi Muhammad Zakaria, 2025. "AI-Powered Early Warning Systems for Emerging Market Crises: Enhancing U.S. Foreign Investment Risk Strategy," Journal of Economics, Finance and Accounting Studies, Al-Kindi Center for Research and Development, vol. 7(5), pages 17-29, August.
  • Handle: RePEc:alk:jefas0:v:7:y:2025:i:5:id:10822
    DOI: 10.32996/jefas.2025.7.5.3
    as

    Download full text from publisher

    File URL: https://al-kindipublisher.com/index.php/jefas/article/view/10822
    File Function: Abstract page
    Download Restriction: no

    File URL: https://al-kindipublisher.com/index.php/jefas/article/download/10822/9604
    File Function: Full text
    Download Restriction: no

    File URL: https://libkey.io/10.32996/jefas.2025.7.5.3?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    More about this item

    Keywords

    ;
    ;
    ;
    ;
    ;
    ;
    ;
    ;
    ;
    ;
    ;
    ;

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:alk:jefas0:v:7:y:2025:i:5:id:10822. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    We have no bibliographic references for this item. You can help adding them by using this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Al-Kindi Center for Research and Development (email available below). General contact details of provider: https://al-kindipublisher.com/index.php/jefas .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.