ResearchSpace

Low default credit scoring using two-class non-parametric kernel density estimation

Show simple item record

dc.contributor.author Rademeyer, E
dc.contributor.author Van der Walt, Christiaan M
dc.contributor.author De Waal, A
dc.date.accessioned 2017-07-28T09:08:43Z
dc.date.available 2017-07-28T09:08:43Z
dc.date.issued 2016-12
dc.identifier.citation Rademeyer, E., Van der Walt, C.M. and De Waal, A. 2016. Low default credit scoring using two-class non-parametric kernel density estimation. Proceedings of the Twenty-Seventh Annual Symposium of the Pattern Recognition Association of South Africa, 30 November - 2 December 2016, Stellenbosch, South Africa. DOI: 10.1109/RoboMech.2016.7813152 en_US
dc.identifier.isbn 978-1-5090-3335-5
dc.identifier.uri http://ieeexplore.ieee.org/document/7813152/
dc.identifier.uri DOI: 10.1109/RoboMech.2016.7813152
dc.identifier.uri http://hdl.handle.net/10204/9369
dc.description Copyright: 2016 IEEE. Due to copyright restrictions, the attached PDF file only contains the abstract of the full text item. For access to the full text item, kindly consult the publisher's website. en_US
dc.description.abstract This paper investigates the performance of two-class classification credit scoring data sets with low default ratios. The standard two-class parametric Gaussian and non-parametric Parzen classifiers are extended, using Bayes’ rule, to include either a class imbalance or a Bernoulli prior. This is done with the aim of addressing the low default probability problem. Furthermore, the performance of Parzen classification with Silverman and Minimum Leave-one-out Entropy (MLE) Gaussian kernel bandwidth estimation is also investigated. en_US
dc.language.iso en en_US
dc.publisher IEEE en_US
dc.relation.ispartofseries Worklist;18147
dc.subject Credit scoring en_US
dc.subject Modelling credit risk en_US
dc.subject Gaussian classifiers en_US
dc.subject Parzen classifiers en_US
dc.title Low default credit scoring using two-class non-parametric kernel density estimation en_US
dc.type Conference Presentation en_US
dc.identifier.apacitation Rademeyer, E., Van der Walt, C. M., & De Waal, A. (2016). Low default credit scoring using two-class non-parametric kernel density estimation. IEEE. http://hdl.handle.net/10204/9369 en_ZA
dc.identifier.chicagocitation Rademeyer, E, Christiaan M Van der Walt, and A De Waal. "Low default credit scoring using two-class non-parametric kernel density estimation." (2016): http://hdl.handle.net/10204/9369 en_ZA
dc.identifier.vancouvercitation Rademeyer E, Van der Walt CM, De Waal A, Low default credit scoring using two-class non-parametric kernel density estimation; IEEE; 2016. http://hdl.handle.net/10204/9369 . en_ZA
dc.identifier.ris TY - Conference Presentation AU - Rademeyer, E AU - Van der Walt, Christiaan M AU - De Waal, A AB - This paper investigates the performance of two-class classification credit scoring data sets with low default ratios. The standard two-class parametric Gaussian and non-parametric Parzen classifiers are extended, using Bayes’ rule, to include either a class imbalance or a Bernoulli prior. This is done with the aim of addressing the low default probability problem. Furthermore, the performance of Parzen classification with Silverman and Minimum Leave-one-out Entropy (MLE) Gaussian kernel bandwidth estimation is also investigated. DA - 2016-12 DB - ResearchSpace DP - CSIR KW - Credit scoring KW - Modelling credit risk KW - Gaussian classifiers KW - Parzen classifiers LK - https://researchspace.csir.co.za PY - 2016 SM - 978-1-5090-3335-5 T1 - Low default credit scoring using two-class non-parametric kernel density estimation TI - Low default credit scoring using two-class non-parametric kernel density estimation UR - http://hdl.handle.net/10204/9369 ER - en_ZA


Files in this item

Thumbnail
Name: Rademeyer_18147_2 ...
Size: 95.50Kb
Format: PDF
Description: Abstract
View/Open

This item appears in the following Collection(s)

Show simple item record