AI-Driven Risk Prediction and Systemic Stability: A Framework for Strengthening U.S. Financial Markets
The American Journal of Applied Sciences · 11.04.2026 · 08 (04) · 18-35
DOI: 10.37547/tajas/volume08issue04-03 ↗Kurz gefasst
Fachartikel in der Zeitschrift The American Journal of Applied Sciences, online erschienen am 11.04.2026. Verfasst von Aftab Uddin.
Abstract (Originalsprache)
Background
Over the last five years, U.S. financial markets have experienced repeated episodes of instability, highlighting the growing importance of detecting algorithmic contagion and systemic risk at an early stage. As automated trading and data-driven financial decision-making expand, explainable artificial intelligence (XAI) offers a promising approach for identifying hidden risk patterns.
Purpose
This study aims to examine how XAI can improve the detection of algorithmic contagion and support systemic risk mitigation in U.S. financial markets.
Methods
The study used a quantitative longitudinal secondary-data design based on five years of daily market data from the SPDR S&P 500 ETF Trust (SPY) and Invesco QQQ Trust (QQQ) from 2020 to 2025. Engineered time-series features supported logistic regression and random forest models, while explainability techniques identified key predictors of contagion-sensitive market states.
Results
Contagion-sensitive periods were concentrated in 2022. In walk-forward validation, logistic regression achieved 0.773 accuracy, 0.311 recall, and 0.175 precision, while random forest achieved 0.821 accuracy, 0.189 recall, and 0.198 precision. In holdout testing, random forest reached 0.905 accuracy but 0.000 recall and precision, highlighting the difficulty of predicting rare stress events.
Conclusion
The study concludes that XAI can strengthen systemic risk monitoring by making machine-learning predictions more transparent and decision-relevant. While predictive performance remains constrained by limited market variables and daily-frequency data, the framework demonstrates strong potential for improving financial surveillance.
Zitieren
APA
Uddin, A. (2026). AI-Driven Risk Prediction and Systemic Stability: A Framework for Strengthening U.S. Financial Markets. The American Journal of Applied Sciences, 08(04), 18-35. https://doi.org/10.37547/tajas/volume08issue04-03
Vancouver
Uddin A. AI-Driven Risk Prediction and Systemic Stability: A Framework for Strengthening U.S. Financial Markets. The American Journal of Applied Sciences. 2026;08(04):18-35. doi:10.37547/tajas/volume08issue04-03
BibTeX
@article{Uddin2026,
author = {Uddin, Aftab},
title = {AI-Driven Risk Prediction and Systemic Stability: A Framework for Strengthening U.S. Financial Markets},
journal = {The American Journal of Applied Sciences},
year = {2026},
volume = {08},
number = {04},
pages = {18--35},
doi = {10.37547/tajas/volume08issue04-03}
}
EUVIMED