Results 1 
6 of
6
Generalized Beta Regression Models for Random LossGivenDefault
, 2008
"... We propose a new framework for modeling systematic risk in LossGivenDefault (LGD) in the context of credit portfolio losses. The class of models is very flexible and accommodates well skewness and heteroscedastic errors. The quantities in the models have simple economic interpretation. Inference o ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
We propose a new framework for modeling systematic risk in LossGivenDefault (LGD) in the context of credit portfolio losses. The class of models is very flexible and accommodates well skewness and heteroscedastic errors. The quantities in the models have simple economic interpretation. Inference of models in this framework can be unified. Moreover, it allows efficient numerical procedures, such as the normal approximation and the saddlepoint approximation, to calculate the portfolio loss distribution, Value at Risk (VaR) and Expected Shortfall (ES).
A Dynamic Hierarchical Bayesian Model for the Probability of Default
, 2010
"... The internal ratingsbased approach under Basel II allows banks to use their internal rating systems to determine capital requirements for credit risk, subject to explicit supervisory approval. The probability of default is one of the key parameters used in the calculation of regulatory capital. All ..."
Abstract
 Add to MetaCart
The internal ratingsbased approach under Basel II allows banks to use their internal rating systems to determine capital requirements for credit risk, subject to explicit supervisory approval. The probability of default is one of the key parameters used in the calculation of regulatory capital. All banks must provide supervisors with an internal estimate of the probability of default. This paper proposes a dynamic hierarchical Bayesian model for the estimation of the probability of default in the presence of missing data, a common occurrence in financial applications. Our framework not only allows us to impute such missing information but also to account for both serial and crosssectional correlations in the estimation of the probability of default. We can also measure the impact of a stress situation on the credit rating. Our model is applied to data on obligors of a domestic bank in South Korea between 2000 and 2003. For comparison, we also consider a hierarchical Bayesian model with a stationary time correlation structure and a classical logistic regression model.
Generalized Beta Regression Models for Random LossGivenDefault
, 2008
"... We propose a new framework for modeling systematic risk in LossGivenDefault (LGD) in the context of credit portfolio losses. The class of models is very flexible and accommodates well skewness and heteroscedastic errors. The quantities in the models have simple economic interpretation. Inference o ..."
Abstract
 Add to MetaCart
We propose a new framework for modeling systematic risk in LossGivenDefault (LGD) in the context of credit portfolio losses. The class of models is very flexible and accommodates well skewness and heteroscedastic errors. The quantities in the models have simple economic interpretation. Inference of models in this framework can be unified. Moreover, it allows efficient numerical procedures, such as the normal approximation and the saddlepoint approximation, to calculate the portfolio loss distribution, Value at Risk (VaR) and Expected Shortfall (ES). 1
credit risk: the shortcomings of the
, 2009
"... Uncertainty in asset correlation for portfolio ..."