Machine-learning models for bankruptcy prediction: do industrial variables matter?

Daniela Bragoli, Camilla Ferretti, Piero Ganugi, Giovanni Marseguerra, Davide Mezzogori, Francesco Zammori

Risultato della ricerca: Contributo in rivistaArticolo in rivistapeer review

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.
Lingua originaleEnglish
pagine (da-a)1-22
Numero di pagine22
RivistaSpatial Economic Analysis
VolumeOctober 2021
DOI
Stato di pubblicazionePubblicato - 2021

Keywords

  • firm distress analysis
  • industrial variables
  • logistic regression
  • machine learning

Fingerprint

Entra nei temi di ricerca di 'Machine-learning models for bankruptcy prediction: do industrial variables matter?'. Insieme formano una fingerprint unica.

Cita questo