Publications

BibTex format

@article{Bellotti:2021:10.1016/j.ijforecast.2020.06.009,
author = {Bellotti, A and Brigo, D and Gambetti, P and Vrins, F},
doi = {10.1016/j.ijforecast.2020.06.009},
journal = {International Journal of Forecasting},
pages = {428--444},
title = {Forecasting recovery rates on non-performing loans with machine learning},
url = {http://dx.doi.org/10.1016/j.ijforecast.2020.06.009},
volume = {37},
year = {2021}
}

Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - We compare the performances of a wide set of regression techniques and machinelearning algorithms for predicting recovery rates on non-performing loans, using a private database from a European debt collection agency. We find that rule-based algorithmssuch as Cubist, boosted trees and random forests perform significantly better than otherapproaches. In addition to loan contract specificities, the predictors referring to the bankrecovery process – prior to the portfolio’s sale to the debt collector – are also proven tostrongly enhance forecasting performances. These variables, derived from the time-series ofcontacts to defaulted clients and clients’ reimbursements to the bank, help all algorithms tobetter identify debtors with different repayment ability and/or commitment, and in generalwith different recovery potential.
AU  - Bellotti,A
AU  - Brigo,D
AU  - Gambetti,P
AU  - Vrins,F
DO  - 10.1016/j.ijforecast.2020.06.009
EP  - 444
PY  - 2021///
SN  - 0169-2070
SP  - 428
TI  - Forecasting recovery rates on non-performing loans with machine learning
T2  - International Journal of Forecasting
UR  - http://dx.doi.org/10.1016/j.ijforecast.2020.06.009
UR  - https://www.sciencedirect.com/science/article/pii/S016920702030100X?via%3Dihub
UR  - http://hdl.handle.net/10044/1/81135
VL  - 37
ER  -

Download

ProfessorDamianoBrigo

Contact

Location

Summary

Citation

BibTex format

RIS format (EndNote, RefMan)