Logit statico, Logit dinamico e modelli hazard

Franco Varetto

CNR-IRCrES, Consiglio Nazionale delle Ricerche, Istituto di Ricerca sulla Crescita Economica Sostenibile, Strada delle Cacce 73, 10135 Torino (TO), Italia

corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

The application of survival analysis to credit risk has received a lot of attention and is the base of many empirical research. Here that analysis has been applied to a sample of Italian corporates working in the metallurgical sector. The survival analysis in the discrete and continuous domains have been compared to the traditional static logit regression. The results have been raised some interesting problems in both life table and dynamic logit regression; in the first case the heterogeneous aggregation of observations can produce difficulties in the interpretation of the metrics of the table; in the second case the treatment of the past observation of bad companies as observation from good companies reduces the accuracy of the model and contributes to obscure the identification of typical patterns to default.

Keywords: Credit Risk, Probability of default, Logistic Regression, Survival Analysis, Cox Proportional Hazard.

JEL codes: C25, C41, G32, G33, L61

DOI: 10.23760/2421-7158.2022.005

How to Cite this Working Paper

Varetto, F. (2022). Logit statico, Logit dinamico e modelli hazard. (CNR-IRCrES Working Paper 5/2022). Torino: Istituto di Ricerca sulla Crescita Economica Sostenibile. Disponibile da http://dx.doi.org/10.23760/2421-7158.2022.005