Bayesian inference for generalized linear mixed models of portfolio credit risk

Alexander J. McNeil; Jonathan P. Wendin

doi:10.1016/j.jempfin.2006.05.002

Bayesian inference for generalized linear mixed models of portfolio credit risk

Alexander J. McNeil, Jonathan P. Wendin^*

^*Corresponding author for this work

Management

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

Abstract

The aims of this paper are threefold. First, we highlight the usefulness of generalized linear mixed models (GLMMs) in the modelling of portfolio credit default risk. The GLMM-setting allows for a flexible specification of the systematic portfolio risk in terms of observed fixed effects and unobserved random effects, in order to explain the phenomena of default dependence and time-inhomogeneity in historical default data. Second, we show that computational Bayesian techniques such as the Gibbs sampler can be successfully applied to fit models with serially correlated random effects, which are special instances of state space models. Third, we provide an empirical study using Standard and Poor's data on U.S. firms. A model incorporating rating category and sector effects, and a macroeconomic proxy variable for state-of-the-economy suggests the presence of a residual, cyclical, latent component in the systematic risk.

Original language	English
Pages (from-to)	131-149
Number of pages	19
Journal	Journal of empirical finance
Volume	14
Issue number	2
DOIs	https://doi.org/10.1016/j.jempfin.2006.05.002
Publication status	Published - Mar 2007

Keywords

Bayesian inference
Credit risk
Generalized linear mixed model
State space model

Access to Document

10.1016/j.jempfin.2006.05.002

Cite this

@article{7b3a29362c304dd9a9f89e74a7babbbd,

title = "Bayesian inference for generalized linear mixed models of portfolio credit risk",

abstract = "The aims of this paper are threefold. First, we highlight the usefulness of generalized linear mixed models (GLMMs) in the modelling of portfolio credit default risk. The GLMM-setting allows for a flexible specification of the systematic portfolio risk in terms of observed fixed effects and unobserved random effects, in order to explain the phenomena of default dependence and time-inhomogeneity in historical default data. Second, we show that computational Bayesian techniques such as the Gibbs sampler can be successfully applied to fit models with serially correlated random effects, which are special instances of state space models. Third, we provide an empirical study using Standard and Poor's data on U.S. firms. A model incorporating rating category and sector effects, and a macroeconomic proxy variable for state-of-the-economy suggests the presence of a residual, cyclical, latent component in the systematic risk.",

keywords = "Bayesian inference, Credit risk, Generalized linear mixed model, State space model",

author = "McNeil, {Alexander J.} and Wendin, {Jonathan P.}",

year = "2007",

month = mar,

doi = "10.1016/j.jempfin.2006.05.002",

language = "English",

volume = "14",

pages = "131--149",

journal = "Journal of empirical finance",

issn = "0927-5398",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - Bayesian inference for generalized linear mixed models of portfolio credit risk

AU - McNeil, Alexander J.

AU - Wendin, Jonathan P.

PY - 2007/3

Y1 - 2007/3

N2 - The aims of this paper are threefold. First, we highlight the usefulness of generalized linear mixed models (GLMMs) in the modelling of portfolio credit default risk. The GLMM-setting allows for a flexible specification of the systematic portfolio risk in terms of observed fixed effects and unobserved random effects, in order to explain the phenomena of default dependence and time-inhomogeneity in historical default data. Second, we show that computational Bayesian techniques such as the Gibbs sampler can be successfully applied to fit models with serially correlated random effects, which are special instances of state space models. Third, we provide an empirical study using Standard and Poor's data on U.S. firms. A model incorporating rating category and sector effects, and a macroeconomic proxy variable for state-of-the-economy suggests the presence of a residual, cyclical, latent component in the systematic risk.

AB - The aims of this paper are threefold. First, we highlight the usefulness of generalized linear mixed models (GLMMs) in the modelling of portfolio credit default risk. The GLMM-setting allows for a flexible specification of the systematic portfolio risk in terms of observed fixed effects and unobserved random effects, in order to explain the phenomena of default dependence and time-inhomogeneity in historical default data. Second, we show that computational Bayesian techniques such as the Gibbs sampler can be successfully applied to fit models with serially correlated random effects, which are special instances of state space models. Third, we provide an empirical study using Standard and Poor's data on U.S. firms. A model incorporating rating category and sector effects, and a macroeconomic proxy variable for state-of-the-economy suggests the presence of a residual, cyclical, latent component in the systematic risk.

KW - Bayesian inference

KW - Credit risk

KW - Generalized linear mixed model

KW - State space model

UR - http://www.scopus.com/inward/record.url?scp=33847255363&partnerID=8YFLogxK

U2 - 10.1016/j.jempfin.2006.05.002

DO - 10.1016/j.jempfin.2006.05.002

M3 - Article

AN - SCOPUS:33847255363

SN - 0927-5398

VL - 14

SP - 131

EP - 149

JO - Journal of empirical finance

JF - Journal of empirical finance

IS - 2

ER -

Bayesian inference for generalized linear mixed models of portfolio credit risk

Abstract

Keywords

Access to Document

Other files and links

Cite this