Modelli di stima della probabilità di default: Il caso di una banca locale
Are you already subscribed?
Login to check
whether this content is already included on your personal or institutional subscription.
Abstract
Based on a sample of 496 firms from the loans portfolio of a local bank and using financial variables we estimate a linear discriminant function for predicting firm default and we used a logistic model to estimate the probability of default. Results highlight that both the models are able to correctly classify over the 95% of sampled firms. However, logistic regression has a better performance in reducing type I error. Therefore, logistic regression represents a useful tool for estimating the probability of default of firms and for improving credit management strategies.
Keywords
- default
- credit scoring
- discriminant analysis
- logistic regression
