Financial Fragility in Emerging Markets: Examining the Innovative Applications of Machine Learning Design Methods

Xiyan Sun, Pei Yuan, Fengge Yao, Zenan Qin, Sijia Yang () and Xiaomei Wang ()
Additional contact information
Xiyan Sun: Yunnan Arts University
Pei Yuan: Henan University of Urban Construction
Fengge Yao: Harbin University of Commerce
Zenan Qin: Harbin University of Commerce
Sijia Yang: Jiangxi Normal University Science and Technology College
Xiaomei Wang: Harbin University of Commerce

Journal of the Knowledge Economy, 2025, vol. 16, issue 2, No 3, 5862-5883

Abstract: Abstract Emerging economies, while exhibiting higher growth rates compared to developed countries, are susceptible to external shocks, leading to financial fragility. Traditional analysis methods often fall short in accuracy and timeliness. This research introduces a novel approach utilizing Back-Propagation Neural Network (BPNN) to predict financial fragility in emerging markets, focusing on the BRICS countries. By considering twelve impactful factors and employing Principal Component Analysis (PCA), five key influencers are identified. The BPNN model is iteratively optimized to achieve superior quality. Historical data validation attests to the model’s effectiveness. The study identifies five critical factors influencing financial fragility: GDP growth rate, inflation rate, monetary policy, interest rates, and bank’s capital-asset ratio. Among these, GDP growth rate emerges as a significant determinant. Positive growth is correlated with financial stability, while a slowdown or negative growth signals elevated risks. Emerging markets are particularly vulnerable to global economic fluctuations due to their reliance on exports and foreign capital. Additionally, weaker financial systems amplify their susceptibility to shocks.The research underscores the importance of building robust financial sectors, replenishing funding buffers, and proactively managing distressed assets in emerging market economies. The proposed BPNN model provides a powerful tool for risk prediction, though it requires strong indicator data support. While computational intensity and interpretability remain challenges, the benefits of BPNNs outweigh these limitations. Effective communication and information exchange across countries and markets are crucial for maintaining stability in emerging market finance. This study contributes valuable insights into the prediction of financial fragility in emerging markets, offering a comprehensive framework for policymakers and financial practitioners to navigate the challenges and opportunities presented by these dynamic economies.

Keywords: Financial fragility; Emerging markets; Neural networks; Machine learning; Risk prediction model (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s13132-023-01731-w Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:jknowl:v:16:y:2025:i:2:d:10.1007_s13132-023-01731-w

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/13132

DOI: 10.1007/s13132-023-01731-w

Access Statistics for this article

Journal of the Knowledge Economy is currently edited by Elias G. Carayannis

More articles in Journal of the Knowledge Economy from Springer, Portland International Center for Management of Engineering and Technology (PICMET)
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().