Machine learning and credit risk:: Empirical evidence from small- and mid-sized businesses

Alessandro Bitetto; Paola Cerchiello; Stefano Filomeni; Alessandra Tanda; Barbara Tarantino

doi:10.1016/j.seps.2023.101746

Machine learning and credit risk: Empirical evidence from small- and mid-sized businesses

Alessandro Bitetto^* (Corresponding Author)
, Paola Cerchiello
, Stefano Filomeni
, Alessandra Tanda
, Barbara Tarantino

^*Corresponding author for this work

Accountancy & Finance, Finance

University of Pavia

Research output: Contribution to journal › Article › peer-review

19 Citations (Scopus)

7 Downloads (Pure)

Abstract

In this paper, we compare two different approaches to estimate the credit risk for small- and mid-sized businesses (SMBs), namely a classic parametric approach, by fitting an ordered probit model, and a non-parametric approach, calibrating a machine learning historical random forest (HRF) model. The models are applied to a unique and proprietary dataset comprising granular firm-level quarterly data collected from a European investment bank and an international insurance company on a sample of 464 Italian SMBs over the period 2015–2017. Results show that the HRF approach outperforms the traditional ordered probit model, highlighting how advanced estimation methodologies that use machine learning techniques can be successfully implemented to predict SMB credit risk, i.e. when facing high asymmetries of information. Moreover, by using Shapley values, we are able to assess the relevance of each variable in predicting SMB credit risk.

Original language	English
Article number	101746
Number of pages	9
Journal	Socio-Economic Planning Sciences
Volume	90
Early online date	8 Nov 2023
DOIs	https://doi.org/10.1016/j.seps.2023.101746
Publication status	Published - Dec 2023

Data Availability Statement

The data that has been used is confidential.

Funding

This research has received funding from the European Union’s Horizon 2020 research and innovation program “PERISCOPE: Pan European Response to the ImpactS of COvid-19 and future Pandemics and Epidemics” , under the Grant Agreement No. 101016233, H2020-SC1-PHE-CORONAVIRUS-2020-2-RTD.

Funders	Funder number
European Research Council	101016233, H2020-SC1-PHE-CORONAVIRUS-2020-2-RTD

Access to Document

10.1016/j.seps.2023.101746Licence: CC BY-NC-ND

Bitetto_etal_SEPS_Machine_Learning_and_Credit_VOR
This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/bync-nd/4.0/).
Final published version, 1.74 MBLicence: CC BY-NC-ND

Cite this

@article{67a5600125fc444e86d4ac6d0e9ffb58,

title = "Machine learning and credit risk:: Empirical evidence from small- and mid-sized businesses",

abstract = "In this paper, we compare two different approaches to estimate the credit risk for small- and mid-sized businesses (SMBs), namely a classic parametric approach, by fitting an ordered probit model, and a non-parametric approach, calibrating a machine learning historical random forest (HRF) model. The models are applied to a unique and proprietary dataset comprising granular firm-level quarterly data collected from a European investment bank and an international insurance company on a sample of 464 Italian SMBs over the period 2015–2017. Results show that the HRF approach outperforms the traditional ordered probit model, highlighting how advanced estimation methodologies that use machine learning techniques can be successfully implemented to predict SMB credit risk, i.e. when facing high asymmetries of information. Moreover, by using Shapley values, we are able to assess the relevance of each variable in predicting SMB credit risk.",

author = "Alessandro Bitetto and Paola Cerchiello and Stefano Filomeni and Alessandra Tanda and Barbara Tarantino",

year = "2023",

month = dec,

doi = "10.1016/j.seps.2023.101746",

language = "English",

volume = "90",

journal = "Socio-Economic Planning Sciences",

issn = "0038-0121",

publisher = "Elsevier Ltd",

}

TY - JOUR

T1 - Machine learning and credit risk:

T2 - Empirical evidence from small- and mid-sized businesses

AU - Bitetto, Alessandro

AU - Cerchiello, Paola

AU - Filomeni, Stefano

AU - Tanda, Alessandra

AU - Tarantino, Barbara

PY - 2023/12

Y1 - 2023/12

N2 - In this paper, we compare two different approaches to estimate the credit risk for small- and mid-sized businesses (SMBs), namely a classic parametric approach, by fitting an ordered probit model, and a non-parametric approach, calibrating a machine learning historical random forest (HRF) model. The models are applied to a unique and proprietary dataset comprising granular firm-level quarterly data collected from a European investment bank and an international insurance company on a sample of 464 Italian SMBs over the period 2015–2017. Results show that the HRF approach outperforms the traditional ordered probit model, highlighting how advanced estimation methodologies that use machine learning techniques can be successfully implemented to predict SMB credit risk, i.e. when facing high asymmetries of information. Moreover, by using Shapley values, we are able to assess the relevance of each variable in predicting SMB credit risk.

AB - In this paper, we compare two different approaches to estimate the credit risk for small- and mid-sized businesses (SMBs), namely a classic parametric approach, by fitting an ordered probit model, and a non-parametric approach, calibrating a machine learning historical random forest (HRF) model. The models are applied to a unique and proprietary dataset comprising granular firm-level quarterly data collected from a European investment bank and an international insurance company on a sample of 464 Italian SMBs over the period 2015–2017. Results show that the HRF approach outperforms the traditional ordered probit model, highlighting how advanced estimation methodologies that use machine learning techniques can be successfully implemented to predict SMB credit risk, i.e. when facing high asymmetries of information. Moreover, by using Shapley values, we are able to assess the relevance of each variable in predicting SMB credit risk.

U2 - 10.1016/j.seps.2023.101746

DO - 10.1016/j.seps.2023.101746

M3 - Article

SN - 0038-0121

VL - 90

JO - Socio-Economic Planning Sciences

JF - Socio-Economic Planning Sciences

M1 - 101746

ER -