The credit rating process and estimation of transition probabilities: A Bayesian approach

Stefanescu, Catalina; Tunaru, Radu; Turnbull, Stuart M.

doi:10.1016/j.jempfin.2008.10.006

Cited by 59 publications

(39 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We propose the use of intensity models to predict for delinquency and default, which have been previously applied to the estimation of rating transitions of corporate loans (see Jarrow et al (1997), Duffie et al (2007), Lando and Skodeberg (2002)), but have not been used on retail loans yet. Also, other approaches for estimating transition probabilities have been applied to corporate loans, for example the standard unobserved latent factor model and Bayesian methods (see for example Stefanescu (2009) and Kadam and Lenk (2008) but in these papers only aggregated data was used.…”

Section: Introductionmentioning

confidence: 99%

Intensity models and transition probabilities for credit card loan delinquencies

Leow

Crook

2014

European Journal of Operational Research

View full text Add to dashboard Cite

Intensity models and transition probabilities for credit card loan delinquenciesCitation for published version: Leow, M & Crook, J 2014, 'Intensity models and transition probabilities for credit card loan delinquencies' European Journal of Operational Research, vol. 236, pp. 685-694. DOI: 10.1016/j.ejor.2013 General rightsCopyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights. Take down policyThe University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer content complies with UK legislation. If you believe that the public display of this file breaches copyright please contact openaccess@ed.ac.uk providing details, and we will remove access to the work immediately and investigate your claim. Intensity models and transition probabilities for credit card loan delinquencies AbstractWe estimate the probability of delinquency and default for a sample of credit card loans using intensity models, via semi-parametric multiplicative hazard models with time-varying covariates. It is the first time these models, previously applied for the estimation of rating transitions, are used on retail loans. Four states are defined in this non-homogenous Markov chain: up-to-date, one month in arrears, two months in arrears, and default; where transitions between states are affected by individual characteristics of the debtor at application and their repayment behaviour since.These intensity estimations allow for insights into the factors that affect movements towards (and recovery from) delinquency, and into default (or not). Results indicate that different types of debtors behave differently while in different states. The probabilities estimated for each type of transition are then used to make out-ofsample predictions over a specified period of time.

show abstract

Section: Introductionmentioning

confidence: 99%

Intensity models and transition probabilities for credit card loan delinquencies

Leow

Crook

2014

European Journal of Operational Research

View full text Add to dashboard Cite

show abstract

“…Overall, the two main technical challenges presented by LDPs are: (i) estimating an accurate PD when no historical defaults are available; and (ii) assessing a model's predictive performance (Stefanescu et al, 2009). Both of these issues arise not only during the validation of the model, but also prior to this, during the construction of the model.…”

Section: The Low-default Portfolio Problem: Previous Workmentioning

confidence: 99%

Using semi-supervised classifiers for credit scoring

Kennedy¹,

Namee²,

Delany³

2013

Journal of the Operational Research Society

View full text Add to dashboard Cite

“…Nous supposons que les probabilités de transition annuelles s'appliquent aussi pour notre période plus courte de six mois. En ce qui concerne les matrices de transition alternatives, consulter, entre autres, Farnsworth et Li (2007), Kiefer et Larson (2007), Feng et al (2008), Frydman et Schuermann (2008 ou encore Stefanescu et al (2009). 16. Il existe plusieurs façons de calculer les différences de marges de l'équation 1 : (i) la marge fixe d'emprunteurs au taux de base moins celle au comptant d'emprunteurs à un taux plus risqué, (ii) la différence entre marges fixes d'emprunteurs au taux de base et à un taux plus risqué, ou (iii) la marge fixe moins la marge au comptant pour chaque classe d'emprunteurs.…”

Section: Implication Principale : De La « Juste » Charge En Capital Aunclassified

Contribution à la réglementation de Bâle-3 : de la consistance interne du continuum du crédit commercial en marquant à la « valeur de modèle » le risque de crédit des engagements de crédit

Château¹

2011

L'Actualité économique

View full text Add to dashboard Cite

Dans le cadre de Bâle-2, l’étude vise à calculer pour les banques canadiennes la charge en capital afférente aux engagements de crédit hors bilan qui s’avère consistante avec celle des prêts au comptant au bilan. Cette consistance s’obtient en sélectionnant le risque de crédit des engagements (par similitude avec celui des prêts commerciaux) au lieu de la durée initiale des engagements (comme prescrit par Bâle-2) comme déterminant principal du calcul. Pour y arriver, nous proposons la « valeur de ou au modèle » comme approche avancée fondée sur les notations internes (NI) : dans celle-ci, le risque de crédit est appréhendé par une option de vente (put) basée sur une distribution de Johnson à quatre paramètres de la « valeur d’endettement à terme » (ce terme désignant la valeur de modèle de l’engagement). Ensuite, lors du calcul de la « juste » charge en capital d’un engagement, le risque « réel » de crédit, évalué comme put à terme de Johnson, est combiné avec la proportion de tirage de la ligne de crédit. Cette charge étant très modérée, l’approche « valeur de modèle » constitue une approche NI avancée « efficiente » puisqu’elle réduit la charge en capital par rapport à celle obtenue par l’approche Bâle-2 simplifiée ou même par rapport à celles obtenues par d’autres procédures NI avancées propres aux banques. Enfin, la valeur au modèle permet de déterminer la charge en capital additionnelle induite par une baisse de la cote de crédit d’un emprunteur. Cette application pratique s’appuie sur la correspondance entre valeurs d’endettement et cotes de crédit d’agences de notation externes ainsi que sur l’existence d’une matrice de probabilités de transition entre notations de crédit en début et fin de période.This research aims to bring some internal consistency to the computation of the credit-risk capital charge of loan commitments and commercial spot loans, namely the components of the commercial credit continuum. Consistency is achieved by using commitment credit risk (as for commercial-loan risk) instead of commitment initial term (as mandated by Basel-2) as the computation main determinant. This is done by proposing marking to model as the bank’s advanced internal-ratings-based (AIRB) approach : credit risk is captured by a four-parameter Johnson distribution of the commitment marked-to-model value (also known as the indebtedness futures value). Once priced as a Johnson’s futures put option, “true” credit risk is combined with the commitment exercise-cum-funding proportion so as to compute the commitment “fair” capital charge. As the latter is quite moderate, marking to model is an “efficient” AIRB approach that offers capital relief with respect to Basel-2 simplified approach but also with respect to the banks’ own alternative AIRB procedures. Marking to model also helps determine the banks’ incremental capital cost caused by borrowers’ credit-rating downgrades. This policy implication hinges on mapping indebtedness values into borrowers’ credit ratings of external agencies and using transition probabilities b...

show abstract

The credit rating process and estimation of transition probabilities: A Bayesian approach

Cited by 59 publications

References 29 publications

Intensity models and transition probabilities for credit card loan delinquencies

Intensity models and transition probabilities for credit card loan delinquencies

Using semi-supervised classifiers for credit scoring

Contribution à la réglementation de Bâle-3 : de la consistance interne du continuum du crédit commercial en marquant à la « valeur de modèle » le risque de crédit des engagements de crédit

Contact Info

Product

Resources

About