Ranking-Based Variable Selection for the Default Risk of Bank Loan Holders

Francesco Giordano (), Marcella Niglio () and Marialuisa Restaino ()
Additional contact information
Francesco Giordano: Università degli Studi di Salerno, Dipartimento di Scienze Economiche e Statistiche
Marcella Niglio: Università degli Studi di Salerno, Dipartimento di Scienze Economiche e Statistiche
Marialuisa Restaino: Università degli Studi di Salerno, Dipartimento di Scienze Economiche e Statistiche

A chapter in Mathematical and Statistical Methods for Actuarial Sciences and Finance, 2022, pp 309-314 from Springer

Abstract: Abstract In this paper we extend the Ranking Based Variable Selection technique (Baranowsky et al., Statistica Sinica 30, 1485–1516 (2020)) to the framework of general linear regression models. After the presentation of the main steps of the algorithm, it is applied to select the variables affecting the repayment ability of bank loan holders. We give evidence that, unlike some largely applied selection methods, the algorithm is robust to the presence of high correlated variables and the number of features selected does not change even when the dataset is contaminated with irrelevant artificial covariates.

Keywords: Screening; Variable selection; High dimension; Ranking (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

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:spr:sprchp:978-3-030-99638-3_50

Ordering information: This item can be ordered from
http://www.springer.com/9783030996383

DOI: 10.1007/978-3-030-99638-3_50

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().