Comparison study of two-step LGD estimation model with probability machines

Yuta Tanoue; Satoshi Yamashita; Hideaki Nagahata

doi:10.1057/s41283-020-00059-y

Comparison study of two-step LGD estimation model with probability machines

Yuta Tanoue, Satoshi Yamashita, Hideaki Nagahata

Research output: Contribution to journal › Article › peer-review

Abstract

Accurate estimation of loss given default is necessary to estimating credit risk. Due to the bi-modal nature of LGD, the two-step LGD estimation model is a promising method for LGD estimation. This study improves the first model in the two-step LGD estimation model using probability machines (random forest, k-nearest neighbors, bagged nearest neighbors, and support vector machines). Furthermore, we compare the predictive performance of each model with traditional logistic regression models. This study confirms that random forest is the best model for developing the first model in the two-step LGD estimation model.

Original language	English
Pages (from-to)	155-177
Number of pages	23
Journal	Risk Management
Volume	22
Issue number	3
DOIs	https://doi.org/10.1057/s41283-020-00059-y
Publication status	Published - 2020 Sep 1
Externally published	Yes

ASJC Scopus subject areas

Business and International Management
Finance
Economics and Econometrics
Strategy and Management

Access to Document

10.1057/s41283-020-00059-y

Fingerprint Dive into the research topics of 'Comparison study of two-step LGD estimation model with probability machines'. Together they form a unique fingerprint.

Cite this

@article{36a4080bb3a4449db88850fe363b2a76,

title = "Comparison study of two-step LGD estimation model with probability machines",

abstract = "Accurate estimation of loss given default is necessary to estimating credit risk. Due to the bi-modal nature of LGD, the two-step LGD estimation model is a promising method for LGD estimation. This study improves the first model in the two-step LGD estimation model using probability machines (random forest, k-nearest neighbors, bagged nearest neighbors, and support vector machines). Furthermore, we compare the predictive performance of each model with traditional logistic regression models. This study confirms that random forest is the best model for developing the first model in the two-step LGD estimation model.",

author = "Yuta Tanoue and Satoshi Yamashita and Hideaki Nagahata",

year = "2020",

month = sep,

day = "1",

doi = "10.1057/s41283-020-00059-y",

language = "English",

volume = "22",

pages = "155--177",

journal = "Risk Management",

issn = "1460-3799",

publisher = "Palgrave Macmillan Ltd.",

number = "3",

}

TY - JOUR

T1 - Comparison study of two-step LGD estimation model with probability machines

AU - Tanoue, Yuta

AU - Yamashita, Satoshi

AU - Nagahata, Hideaki

PY - 2020/9/1

Y1 - 2020/9/1

N2 - Accurate estimation of loss given default is necessary to estimating credit risk. Due to the bi-modal nature of LGD, the two-step LGD estimation model is a promising method for LGD estimation. This study improves the first model in the two-step LGD estimation model using probability machines (random forest, k-nearest neighbors, bagged nearest neighbors, and support vector machines). Furthermore, we compare the predictive performance of each model with traditional logistic regression models. This study confirms that random forest is the best model for developing the first model in the two-step LGD estimation model.

AB - Accurate estimation of loss given default is necessary to estimating credit risk. Due to the bi-modal nature of LGD, the two-step LGD estimation model is a promising method for LGD estimation. This study improves the first model in the two-step LGD estimation model using probability machines (random forest, k-nearest neighbors, bagged nearest neighbors, and support vector machines). Furthermore, we compare the predictive performance of each model with traditional logistic regression models. This study confirms that random forest is the best model for developing the first model in the two-step LGD estimation model.

UR - http://www.scopus.com/inward/record.url?scp=85084090714&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85084090714&partnerID=8YFLogxK

U2 - 10.1057/s41283-020-00059-y

DO - 10.1057/s41283-020-00059-y

M3 - Article

AN - SCOPUS:85084090714

VL - 22

SP - 155

EP - 177

JO - Risk Management

JF - Risk Management

SN - 1460-3799

IS - 3

ER -