Forecasting Recovery Rates on Non-Performing Loans with Machine Learning

Credit Scoring and Credit Control Conference XVI

29 Pages Posted: 12 Aug 2019 Last revised: 6 Sep 2019

See all articles by Anthony Bellotti

Anthony Bellotti

University of Nottingham Ningbo China

Damiano Brigo

Imperial College London - Department of Mathematics

Paolo Gambetti

Louvain Finance Center (LFIN), UCLouvain

Frédéric D. Vrins

LFIN/LIDAM, UCLouvain

Date Written: August 29, 2019

Abstract

We compare the performances of a wide set of regression techniques and machine learning algorithms for predicting recovery rates on non-performing loans, using a private database from a European debt collection agency. We find that rule-based algorithms such as Cubist, boosted trees and random forests perform significantly better than other approaches. In addition to loan contract specificities, the predictors referring to the bank recovery process -- prior to the portfolio's sale to the debt collector -- are also proven to strongly enhance forecasting performances. These variables, derived from the time-series of contacts to defaulted clients and clients' reimbursements to the bank, help all algorithms to better identify debtors with different repayment ability and/or commitment, and in general with different recovery potential.

Keywords: Risk Management, Recovery Rate, Non-Performing Loans, Forecasting

JEL Classification: G21, G23, C52

Suggested Citation

Bellotti, Anthony and Brigo, Damiano and Gambetti, Paolo and Vrins, Frederic Daniel, Forecasting Recovery Rates on Non-Performing Loans with Machine Learning (August 29, 2019). Credit Scoring and Credit Control Conference XVI, Available at SSRN: https://ssrn.com/abstract=3434412 or http://dx.doi.org/10.2139/ssrn.3434412

Anthony Bellotti

University of Nottingham Ningbo China ( email )

199 Taikang East Road
Yinzhou
Ningbo, Zhejiang 315100
China
+1008619883003047 (Phone)

Damiano Brigo

Imperial College London - Department of Mathematics ( email )

South Kensington Campus
London SW7 2AZ, SW7 2AZ
United Kingdom

HOME PAGE: http://www.imperial.ac.uk/people/damiano.brigo

Paolo Gambetti (Contact Author)

Louvain Finance Center (LFIN), UCLouvain ( email )

34 Voie du Roman Pays - L1.03.01
Louvain-la-Neuve, 1348
Belgium

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
628
Abstract Views
2,173
Rank
77,862
PlumX Metrics