Journal of
Italian
Banking
Association

year 100
 

Archive » January 2023 » Early warning in the Shipping sector: a comparison between traditional and Machine Learning models

Early warning in the Shipping sector: a comparison between traditional and Machine Learning models

Luca Giordano, Gianluca Mattarocci
January 2023 - n. 1
Keywords: Early warning, probabilità di default, shipping
Jel codes: G33, G21, L91

The pro-active credit risk approach aims to identify variables that allow to anticipate borrowers' default and literature has shown that the best proxies differ sector by sector. The Shipping industry is one of the most complex for forecasting insolvency, and an effective early warning system requires to consider not only the financial statement fundamentals but also of the business model and its sustainability in the medium and long term. The article analyzes a representative dataset of Italian companies operating in the maritime freight transportation sector in the last nine years, and it proposes a method for developing an early warning system. Results show that data related to debt, profitability, asset structure are necessary, together with specific Shipping parameters. Machine learning solutions allow a better early warning system with respect to traditional procedures.

Interested in this paper?
Buy the issue