Domain Knowledge Features versus LASSO Features in Predicting Risk of Corporate Bankruptcy—DEA Approach

Martina Mokrišová () and Jarmila Horváthová
Additional contact information
Martina Mokrišová: Faculty of Management and Business, University of Prešov, Konštantínova 16, 080 01 Prešov, Slovakia
Jarmila Horváthová: Faculty of Management and Business, University of Prešov, Konštantínova 16, 080 01 Prešov, Slovakia

Risks, 2023, vol. 11, issue 11, 1-18

Abstract: Predicting the risk of corporate bankruptcy is one of the most important challenges for researchers dealing with the issue of financial health evaluation. The risk of corporate bankruptcy is most often assessed with the use of early warning models. The results of these models are significantly influenced by the financial features entering them. The aim of this paper was to select the most suitable financial features for bankruptcy prediction. The research sample consisted of enterprises conducting a business within the Slovak construction industry. The features were selected using the domain knowledge (DK) approach and Least Absolute Shrinkage and Selection Operator (LASSO). The performance of VRS DEA (Variable Returns to Scale Data Envelopment Analysis) models was assessed with the use of accuracy, ROC (Receiver Operating Characteristics) curve, AUC (Area Under the Curve) and Somers’ D. The results show that the DK+DEA model achieved slightly better AUC and Somers’ D compared to the LASSO+DEA model. On the other hand, the LASSO+DEA model shows a smaller deviation in the number of identified businesses on the financial distress frontier. The added value of this research is the finding that the application of DK features achieves significant results in predicting businesses’ bankruptcy. The added value for practice is the selection of predictors of bankruptcy for the analyzed sample of enterprises.

Keywords: data envelopment analysis; domain knowledge; feature; LASSO; risk of corporate bankruptcy (search for similar items in EconPapers)
JEL-codes: C G0 G1 G2 G3 K2 M2 M4 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-9091/11/11/199/pdf (application/pdf)
https://www.mdpi.com/2227-9091/11/11/199/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jrisks:v:11:y:2023:i:11:p:199-:d:1280888

Access Statistics for this article

Risks is currently edited by Mr. Claude Zhang

More articles in Risks from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().