Loss given default estimation: A two-stage model with classification tree-based boosting and support vector logistic regression

Yuta Tanoue, Satoshi Yamashita

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The Basel Accords allow banks to estimate credit risk. Accordingly, more attention has been dedicated recently to the analysis of loss given default (LGD) and the development of an LGD estimation model. In this study, using a data set composed of five Japanese regional banks, we propose an LGD estimation model using a twostage model, classification tree-based boosting and support vector regression (SVR). We compare the proposed model’s predictive performance with existing models by performing cross-validation and out-of-time validation. As a result, we find that incorporating nonlinearity into the LGD estimation model by classification and SVR improves its predictive performance. Further, we confirm that the boosting method improves the model’s predictive performance.

Original languageEnglish
Pages (from-to)19-37
Number of pages19
JournalJournal of Risk
Volume21
Issue number4
DOIs
Publication statusPublished - 2019 Apr
Externally publishedYes

Keywords

  • Boosting
  • Credit risk
  • Loss given default (LGD)
  • Platt scaling
  • Risk management

ASJC Scopus subject areas

  • Finance
  • Strategy and Management

Fingerprint Dive into the research topics of 'Loss given default estimation: A two-stage model with classification tree-based boosting and support vector logistic regression'. Together they form a unique fingerprint.

  • Cite this