A hybrid information approach to predict corporate credit risk

Di Bu, Simone Kelly, Yin Liao, Qing Zhou

Research output: Contribution to journalArticle

Abstract

This study proposes a hybrid information approach to predict corporate credit risk. In contrast to the previous literature that debates which credit risk model is the best, we pool information from a diverse set of structural and reduced-form models to produce a model combination based on credit risk prediction. Compared with each single model, the pooled strategies yield consistently lower average risk prediction errors over time. We also find that while the reduced-form models contribute more in the pooled strategies for speculative-grade names and longer maturities, the structural models have higher weights for shorter maturities and investment grade names.

LanguageEnglish
Pages1062-1078
Number of pages17
JournalJournal of Futures Markets
Volume38
Issue number9
Early online date21 May 2018
DOIs
Publication statusPublished - Sep 2018

Fingerprint

Reduced-form model
Maturity
Credit risk
Prediction error
Structural model
Prediction
Credit risk models

Cite this

Bu, Di ; Kelly, Simone ; Liao, Yin ; Zhou, Qing. / A hybrid information approach to predict corporate credit risk. In: Journal of Futures Markets. 2018 ; Vol. 38, No. 9. pp. 1062-1078.
@article{7701410ada274b9f9bb71e7644f1895f,
title = "A hybrid information approach to predict corporate credit risk",
abstract = "This study proposes a hybrid information approach to predict corporate credit risk. In contrast to the previous literature that debates which credit risk model is the best, we pool information from a diverse set of structural and reduced-form models to produce a model combination based on credit risk prediction. Compared with each single model, the pooled strategies yield consistently lower average risk prediction errors over time. We also find that while the reduced-form models contribute more in the pooled strategies for speculative-grade names and longer maturities, the structural models have higher weights for shorter maturities and investment grade names.",
author = "Di Bu and Simone Kelly and Yin Liao and Qing Zhou",
year = "2018",
month = "9",
doi = "10.1002/fut.21930",
language = "English",
volume = "38",
pages = "1062--1078",
journal = "Journal of Futures Markets",
issn = "0270-7314",
publisher = "Wiley-Academy",
number = "9",

}

A hybrid information approach to predict corporate credit risk. / Bu, Di; Kelly, Simone; Liao, Yin; Zhou, Qing.

In: Journal of Futures Markets, Vol. 38, No. 9, 09.2018, p. 1062-1078.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A hybrid information approach to predict corporate credit risk

AU - Bu, Di

AU - Kelly, Simone

AU - Liao, Yin

AU - Zhou, Qing

PY - 2018/9

Y1 - 2018/9

N2 - This study proposes a hybrid information approach to predict corporate credit risk. In contrast to the previous literature that debates which credit risk model is the best, we pool information from a diverse set of structural and reduced-form models to produce a model combination based on credit risk prediction. Compared with each single model, the pooled strategies yield consistently lower average risk prediction errors over time. We also find that while the reduced-form models contribute more in the pooled strategies for speculative-grade names and longer maturities, the structural models have higher weights for shorter maturities and investment grade names.

AB - This study proposes a hybrid information approach to predict corporate credit risk. In contrast to the previous literature that debates which credit risk model is the best, we pool information from a diverse set of structural and reduced-form models to produce a model combination based on credit risk prediction. Compared with each single model, the pooled strategies yield consistently lower average risk prediction errors over time. We also find that while the reduced-form models contribute more in the pooled strategies for speculative-grade names and longer maturities, the structural models have higher weights for shorter maturities and investment grade names.

UR - http://www.scopus.com/inward/record.url?scp=85047487122&partnerID=8YFLogxK

U2 - 10.1002/fut.21930

DO - 10.1002/fut.21930

M3 - Article

VL - 38

SP - 1062

EP - 1078

JO - Journal of Futures Markets

T2 - Journal of Futures Markets

JF - Journal of Futures Markets

SN - 0270-7314

IS - 9

ER -