Bankruptcy Prediction of Small and Medium Enterprises Using a Flexible Binary Generalized Extreme Value Model

TL;DR

This paper presents a flexible binary regression model for SME bankruptcy prediction that adapts to data-driven relationships, outperforming traditional logistic models in validation tests.

Contribution

It introduces a novel generalized extreme value-based link function that relaxes common assumptions, improving predictive accuracy for SME default risk.

Findings

Outperforms logistic models in validation tests

Flexible modeling of covariate effects improves prediction

Effective for different default horizons

Abstract

We introduce a binary regression accounting-based model for bankruptcy prediction of small and medium enterprises (SMEs). The main advantage of the model lies in its predictive performance in identifying defaulted SMEs. Another advantage, which is especially relevant for banks, is that the relationship between the accounting characteristics of SMEs and response is not assumed a priori (e.g., linear, quadratic or cubic) and can be determined from the data. The proposed approach uses the quantile function of the generalized extreme value distribution as link function as well as smooth functions of accounting characteristics to flexibly model covariate effects. Therefore, the usual assumptions in scoring models of symmetric link function and linear or pre-specied covariate-response relationships are relaxed. Out-of-sample and out-of-time validation on Italian data shows that our proposal outperforms the commonly used (logistic) scoring model for different default horizons.