A Cyclical Model of Multiple-Horizon Credit Rating Transitions and Default

Metz, Albert

doi:10.2139/ssrn.993474

Cited by 2 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The rating dynamics variables relate to up‐ or downgrade behavior and as such capture the rating momentum behavior. Generally, a counterparty that was downgraded recently will have a greater probability to be downgraded again than to be upgraded, and vice versa (Metz and Cantor, ).…”

Section: Resultsmentioning

confidence: 99%

“…Moreover, to avoid overfitting and undue model complexity, a Bayesian information criterion (BIC) stopping criterion is adopted. It is defined using the model deviance, and the variable selection procedure is stopped when no extra variables can be selected that provide an improvement in model deviance over Dev as defined by equation (10), where N obs represents the number of observations (see, for example, Neter et al, 1996):…”

Section: Model Estimationmentioning

confidence: 99%

“…whether a rating previously has been downgraded or upgraded. Rating dynamics have been shown in the literature to effectively influence migration behavior (Metz and Cantor, ).…”

Section: Literature Reviewmentioning

confidence: 99%

See 2 more Smart Citations

A Novel Credit Rating Migration Modeling Approach Using Macroeconomic Indicators

Berteloot¹,

Verbeke

Castermans³

et al. 2013

Journal of Forecasting

View full text Add to dashboard Cite

Modeling credit rating migrations conditional on macroeconomic conditions allows financial institutions to assess, analyze, and manage the risk related to a credit portfolio. Existing methodologies to model credit rating migrations conditional on the business cycle suffer from poor accuracy, difficult readability, or model inconsistencies. The modeling methodology proposed in this paper extends ordinal logistic regression to estimate the complete migration matrix including default rates as a function of rating dynamics and macroeconomic indicators. The gradient and Hessian derivations show efficient optimization within the Levenberg–Marquardt algorithm. The proposed modeling methodology is applied to model corporate rating migrations using historical data from 1984 to 2011. It is shown that the resulting model captures the cyclical behavior of credit rating migrations and default rates, and is able to approximate historic migration levels with good precision. The model therefore permits analysis of the impact of economical downturn conditions on a credit portfolio. Copyright © 2013 John Wiley & Sons, Ltd.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Model Estimationmentioning

confidence: 99%