Forecasting Loan Default in Europe with Machine Learning

Abstract We use a dataset of 12 million residential mortgages to investigate the loan default behavior in several European countries. We model the default occurrence as a function of borrower characteristics, loan-specific variables, and local economic conditions. We compare the performance of a set...

Full description

Saved in:
Bibliographic Details
Published in:Journal of financial econometrics 2023-03, Vol.21 (2), p.569-596
Main Authors: Barbaglia, Luca, Manzan, Sebastiano, Tosetti, Elisa
Format: Article
Language:English
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c337t-2964662ac77f13ce541c66e3a85c6fa11752e4341453bb9cc7efd84c4d8a655c3
cites cdi_FETCH-LOGICAL-c337t-2964662ac77f13ce541c66e3a85c6fa11752e4341453bb9cc7efd84c4d8a655c3
container_end_page 596
container_issue 2
container_start_page 569
container_title Journal of financial econometrics
container_volume 21
creator Barbaglia, Luca
Manzan, Sebastiano
Tosetti, Elisa
description Abstract We use a dataset of 12 million residential mortgages to investigate the loan default behavior in several European countries. We model the default occurrence as a function of borrower characteristics, loan-specific variables, and local economic conditions. We compare the performance of a set of machine learning algorithms relative to the logistic regression, finding that they perform significantly better in providing predictions. The most important variables in explaining loan default are the interest rate and the local economic characteristics. The existence of relevant geographical heterogeneity in the variable importance points at the need for regionally tailored risk-assessment policies in Europe.
doi_str_mv 10.1093/jjfinec/nbab010
format article
fullrecord <record><control><sourceid>oup_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1093_jjfinec_nbab010</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><oup_id>10.1093/jjfinec/nbab010</oup_id><sourcerecordid>10.1093/jjfinec/nbab010</sourcerecordid><originalsourceid>FETCH-LOGICAL-c337t-2964662ac77f13ce541c66e3a85c6fa11752e4341453bb9cc7efd84c4d8a655c3</originalsourceid><addsrcrecordid>eNqFjzFPwzAUhC0EEqUws3pGCvWLHTsZUWkLUhALzNHLq00TFSeyEyH-PUGNWJnuhvtO-hi7BXEPopCrtnWNt7TyNdYCxBlbgDJFkisw539dFJfsKsZWiFSnChYs33bBEsah8R-87NDzR-twPA688Xwzhq63_KsZDvwF6TD989Ji8NP4ml04PEZ7M-eSvW83b-unpHzdPa8fyoSkNEOSFlppnSIZ40CSzRSQ1lZinpF2CGCy1CqpQGWyrgsiY90-V6T2OeosI7lkq9MvhS7GYF3Vh-YTw3cFovoVr2bxahafiLsT0Y39v-MfINtcfw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Forecasting Loan Default in Europe with Machine Learning</title><source>EconLit s plnými texty</source><source>Oxford Journals Online</source><creator>Barbaglia, Luca ; Manzan, Sebastiano ; Tosetti, Elisa</creator><creatorcontrib>Barbaglia, Luca ; Manzan, Sebastiano ; Tosetti, Elisa</creatorcontrib><description>Abstract We use a dataset of 12 million residential mortgages to investigate the loan default behavior in several European countries. We model the default occurrence as a function of borrower characteristics, loan-specific variables, and local economic conditions. We compare the performance of a set of machine learning algorithms relative to the logistic regression, finding that they perform significantly better in providing predictions. The most important variables in explaining loan default are the interest rate and the local economic characteristics. The existence of relevant geographical heterogeneity in the variable importance points at the need for regionally tailored risk-assessment policies in Europe.</description><identifier>ISSN: 1479-8409</identifier><identifier>EISSN: 1479-8417</identifier><identifier>DOI: 10.1093/jjfinec/nbab010</identifier><language>eng</language><publisher>Oxford University Press</publisher><ispartof>Journal of financial econometrics, 2023-03, Vol.21 (2), p.569-596</ispartof><rights>The Author(s) 2021. Published by Oxford University Press. 2021</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c337t-2964662ac77f13ce541c66e3a85c6fa11752e4341453bb9cc7efd84c4d8a655c3</citedby><cites>FETCH-LOGICAL-c337t-2964662ac77f13ce541c66e3a85c6fa11752e4341453bb9cc7efd84c4d8a655c3</cites><orcidid>0000-0001-5930-5392</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,776,780,27901,27902</link.rule.ids></links><search><creatorcontrib>Barbaglia, Luca</creatorcontrib><creatorcontrib>Manzan, Sebastiano</creatorcontrib><creatorcontrib>Tosetti, Elisa</creatorcontrib><title>Forecasting Loan Default in Europe with Machine Learning</title><title>Journal of financial econometrics</title><description>Abstract We use a dataset of 12 million residential mortgages to investigate the loan default behavior in several European countries. We model the default occurrence as a function of borrower characteristics, loan-specific variables, and local economic conditions. We compare the performance of a set of machine learning algorithms relative to the logistic regression, finding that they perform significantly better in providing predictions. The most important variables in explaining loan default are the interest rate and the local economic characteristics. The existence of relevant geographical heterogeneity in the variable importance points at the need for regionally tailored risk-assessment policies in Europe.</description><issn>1479-8409</issn><issn>1479-8417</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>TOX</sourceid><recordid>eNqFjzFPwzAUhC0EEqUws3pGCvWLHTsZUWkLUhALzNHLq00TFSeyEyH-PUGNWJnuhvtO-hi7BXEPopCrtnWNt7TyNdYCxBlbgDJFkisw539dFJfsKsZWiFSnChYs33bBEsah8R-87NDzR-twPA688Xwzhq63_KsZDvwF6TD989Ji8NP4ml04PEZ7M-eSvW83b-unpHzdPa8fyoSkNEOSFlppnSIZ40CSzRSQ1lZinpF2CGCy1CqpQGWyrgsiY90-V6T2OeosI7lkq9MvhS7GYF3Vh-YTw3cFovoVr2bxahafiLsT0Y39v-MfINtcfw</recordid><startdate>20230330</startdate><enddate>20230330</enddate><creator>Barbaglia, Luca</creator><creator>Manzan, Sebastiano</creator><creator>Tosetti, Elisa</creator><general>Oxford University Press</general><scope>TOX</scope><scope>AAYXX</scope><scope>CITATION</scope><orcidid>https://orcid.org/0000-0001-5930-5392</orcidid></search><sort><creationdate>20230330</creationdate><title>Forecasting Loan Default in Europe with Machine Learning</title><author>Barbaglia, Luca ; Manzan, Sebastiano ; Tosetti, Elisa</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c337t-2964662ac77f13ce541c66e3a85c6fa11752e4341453bb9cc7efd84c4d8a655c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Barbaglia, Luca</creatorcontrib><creatorcontrib>Manzan, Sebastiano</creatorcontrib><creatorcontrib>Tosetti, Elisa</creatorcontrib><collection>Oxford Academic Journals (Open Access)</collection><collection>CrossRef</collection><jtitle>Journal of financial econometrics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Barbaglia, Luca</au><au>Manzan, Sebastiano</au><au>Tosetti, Elisa</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Forecasting Loan Default in Europe with Machine Learning</atitle><jtitle>Journal of financial econometrics</jtitle><date>2023-03-30</date><risdate>2023</risdate><volume>21</volume><issue>2</issue><spage>569</spage><epage>596</epage><pages>569-596</pages><issn>1479-8409</issn><eissn>1479-8417</eissn><abstract>Abstract We use a dataset of 12 million residential mortgages to investigate the loan default behavior in several European countries. We model the default occurrence as a function of borrower characteristics, loan-specific variables, and local economic conditions. We compare the performance of a set of machine learning algorithms relative to the logistic regression, finding that they perform significantly better in providing predictions. The most important variables in explaining loan default are the interest rate and the local economic characteristics. The existence of relevant geographical heterogeneity in the variable importance points at the need for regionally tailored risk-assessment policies in Europe.</abstract><pub>Oxford University Press</pub><doi>10.1093/jjfinec/nbab010</doi><tpages>28</tpages><orcidid>https://orcid.org/0000-0001-5930-5392</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1479-8409
ispartof Journal of financial econometrics, 2023-03, Vol.21 (2), p.569-596
issn 1479-8409
1479-8417
language eng
recordid cdi_crossref_primary_10_1093_jjfinec_nbab010
source EconLit s plnými texty; Oxford Journals Online
title Forecasting Loan Default in Europe with Machine Learning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-03T09%3A47%3A20IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-oup_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Forecasting%20Loan%20Default%20in%20Europe%20with%20Machine%20Learning&rft.jtitle=Journal%20of%20financial%20econometrics&rft.au=Barbaglia,%20Luca&rft.date=2023-03-30&rft.volume=21&rft.issue=2&rft.spage=569&rft.epage=596&rft.pages=569-596&rft.issn=1479-8409&rft.eissn=1479-8417&rft_id=info:doi/10.1093/jjfinec/nbab010&rft_dat=%3Coup_cross%3E10.1093/jjfinec/nbab010%3C/oup_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c337t-2964662ac77f13ce541c66e3a85c6fa11752e4341453bb9cc7efd84c4d8a655c3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_oup_id=10.1093/jjfinec/nbab010&rfr_iscdi=true