Machine-Learning Based Default Prediction: The Role of External Audits and Financial Constraints

Hyunjun Song, Dojoon Park, Zoonky Lee and Yong Joo Kang
Additional contact information
Hyunjun Song: Ajou University, South Korea
Dojoon Park: Kongju National University, South Korea
Zoonky Lee: Yonsei University, South Korea
Yong Joo Kang: Thompson Rivers University, Canada

Journal of Economic Development, 2026, vol. 51, issue 2, 13-41

Abstract: We investigate the impact of external audits and financial constraints on predicting small and medium-sized enterprise (SME) defaults in Korea using data for both externally and non-externally audited firms. The importance of a SME-specific default model is highlighted through the performance comparison of default prediction models for SMEs and large firms. Since SMEs are more vulnerable to financial market shocks, variables related to cash reserves and financial constraints are included. The analysis by feature importance shows that variables related to cash balance and financial constraint contribute towards an improvement in default prediction for Korean SMEs.

Keywords: Default Prediction; Small and Medium Enterprises; External Audits; Financial Constraints; Machine Learning (search for similar items in EconPapers)
JEL-codes: C19 C53 C55 C65 G17 (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://jed.cau.ac.kr/archives/51-2/51-2-2.pdf full text
None

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:ris:jecdev:023036

DOI: 10.35866/caujed.2026.51.2.002

Access Statistics for this article

Journal of Economic Development is currently edited by Sung Y. Park

More articles in Journal of Economic Development from The Economic Research Institute, Chung-Ang University Room 1137, Building 310, Chung-Ang University, 84 Heukseok-ro, Dongjak-gu, Seoul 06974, South Korea. Contact information at EDIRC.
Bibliographic data for series maintained by Jo A Jeong ().