Practical insights into predicting defaults in small and medium-sized enterprises
Year: 2024
Type: article
Abstract: This article provides practical insights into predicting defaults in small and medium-sized enterprises (SMEs). Building on a comprehensive review and empirical evaluation of methodologies from two detailed studies, we highlight the most effective estimation methods, feature selection techniques, and validation approaches. Results show that machine learning models, particularly Light Gradient Boosting Machine and Extreme Gradient Boosting, offer superior predictive accuracy. Additionally, proper validation and feature selection are critical to improving performance. We offer actionable recommendations for practitioners and policy makers to enhance decision-making processes, support SME growth, and mitigate financial risks, ultimately contributing to economic stability and development.
Authors: Hamid Cheraghali, Peter Molnár
Language: English
Subfield: Accounting
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4404954820