A Spatio-Temporal Machine Learning Model for Mortgage Credit Risk: Default Probabilities and Loan Portfolios

TL;DR

This paper presents a new machine learning model that combines tree-boosting with a spatio-temporal Gaussian process to improve mortgage credit risk prediction by capturing complex interactions and unobserved spatio-temporal effects.

Contribution

The paper introduces a novel combined model for credit risk that effectively captures non-linearities, interactions, and spatio-temporal frailty effects, improving prediction accuracy.

Findings

More accurate default probability predictions than traditional models

Enhanced loan portfolio loss distribution forecasts

Identification of key non-linear and spatio-temporal effects

Abstract

We introduce a novel machine learning model for credit risk by combining tree-boosting with a latent spatio-temporal Gaussian process model accounting for frailty correlation. This allows for modeling non-linearities and interactions among predictor variables in a flexible data-driven manner and for accounting for spatio-temporal variation that is not explained by observable predictor variables. We also show how estimation and prediction can be done in a computationally efficient manner. In an application to a large U.S. mortgage credit risk data set, we find that both predictive default probabilities for individual loans and predictive loan portfolio loss distributions obtained with our novel approach are more accurate compared to conventional independent linear hazard models and also linear spatio-temporal models. Using interpretability tools for machine learning models, we find that the likely reasons for this outperformance are strong interaction and non-linear effects in the predictor variables and the presence of spatio-temporal frailty effects.