Machine-learning models for bankruptcy prediction: do industrial variables matter?. Issue 2 (3rd April 2022)

• Record Type: Journal Article
• Authors: Bragoli, Daniela; Ferretti, Camilla; Ganugi, Piero; Marseguerra, Giovanni; Mezzogori, Davide; Zammori, Francesco
• Is Part Of: Spatial economic analysis. Volume 17:Issue 2(2022)
• Page Start: 156
• Page End: 177
• Publication Date: 2022-04-03
• Subjects: firm distress analysis -- machine learning -- logistic regression -- industrial variables; C45 -- C52 -- G33 -- R11; Space in economics -- Periodicals; Economic geography -- Periodicals; Space in economics -- Econometric models -- Periodicals; Urbanization -- Periodicals; Espace (Économie politique) -- Périodiques; Économétrie -- Périodiques; Géographie économique -- Périodiques; Économie industrielle -- Périodiques; Économie du travail -- Périodiques; Économie urbaine -- Périodiques; Politique économique -- Périodiques; Relations économiques internationales -- Périodiques; 330.9
• DOI: 10.1080/17421772.2021.1977377
• Languages: English
• ISSNs: 1742-1772

Abstract

ABSTRACT: We provide a predictive model specifically designed for the Italian economy that classifies solvent and insolvent firms one year in advance using the AIDA Bureau van Dijk data set for the period 2007–15. We apply a full battery of bankruptcy forecasting models, including both traditional and more sophisticated machine-learning techniques, and add to the financial ratios used in the literature a set of industrial/regional variables. We find that XGBoost is the best performer, and that industrial/regional variables are important. Moreover, belonging to a district, having a high mark-up and a greater market share diminish bankruptcy probability.