Application of Classification Trees for Credit Rating Prediction

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27510%2F18%3A10239529" target="_blank" >RIV/61989100:27510/18:10239529 - isvavai.cz</a>
Result on the web
<a href="https://www.ekf.vsb.cz/export/sites/ekf/rmfr/en/Conference_proceedings/Proceedings/Part_I_finalni.pdf" target="_blank" >https://www.ekf.vsb.cz/export/sites/ekf/rmfr/en/Conference_proceedings/Proceedings/Part_I_finalni.pdf</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Application of Classification Trees for Credit Rating Prediction
Original language description
This paper is focused on credit rating modelling using the technique of recursive partitioning (CART). In this study, we apply the methods of decision tree learning to create models for credit rating prediction. The models can be used to assess and monitor credit quality of companies and determine the contribution of financial predictors, especially when a certified rating is not available. In this study we present one nominal and four ordinal models based upon different splitting and pruning rules. The focus is paid to the performance comparison of various algorithms for this ordinal classification problem in terms of Somers' delta and ROC curves. Main findings of our study suggest that for our data both approaches perform well and provide comparable results in terms of classification ability.
Czech name
—
Czech description
—

Project
<a href="/en/project/EE2.3.20.0296" target="_blank" >EE2.3.20.0296: Research team for modelling of economic and financial processes at VSB-TU Ostrava</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Managing and Modelling of Financial Risks : 9th international scientific conference : 5th-6th September 2018, Ostrava, Czech Republic : proceedings
ISBN
978-80-248-4225-7
ISSN
2464-6970
e-ISSN
2464-6989
Number of pages
8
Pages from-to
121-128
Publisher name
VŠB - Technical University of Ostrava
Place of publication
Ostrava
Event location
Ostrava
Event date
Sep 5, 2018
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)