Grabit: Gradient Tree-Boosted Tobit Models for Default Prediction

TL;DR

The paper introduces Grabit, a novel gradient tree-boosted Tobit model designed to improve default prediction accuracy in imbalanced datasets, demonstrated on Swiss SME loan data.

Contribution

It presents a new hybrid model combining gradient boosting with Tobit models to enhance default prediction in imbalanced scenarios.

Findings

Significant improvement over existing methods in predictive accuracy.

Effective leverage of auxiliary data for better predictions.

Applicable to real-world SME default prediction tasks.

Abstract

A frequent problem in binary classification is class imbalance between a minority and a majority class such as defaults and non-defaults in default prediction. In this article, we introduce a novel binary classification model, the Grabit model, which is obtained by applying gradient tree boosting to the Tobit model. We show how this model can leverage auxiliary data to obtain increased predictive accuracy for imbalanced data. We apply the Grabit model to predicting defaults on loans made to Swiss small and medium-sized enterprises (SME) and obtain a large and significant improvement in predictive performance compared to other state-of-the-art approaches.