Portfolio credit risk models

Greg M Gupton

There are several portfolio-level methodologies for estimating a portfolio’s CreditVAR (value-at-risk due to credit quality changes). The natural question is, “which framework is best?” In a very significant way, this is driven by the data that are available. If the best data for evaluating credit quality are categorical credit ratings (either from credit rating agencies or a bank’s own internal ratings), then the best models of credit risk will use these as input, likely apply Markov analysis, and end up looking like CreditMetrics or CreditPortfolioView. If the best (most timely and impartial) credit quality inferences can be extracted from equity prices, then the best models will use these, probably apply a Merton-type structural approach and end up looking like PortfolioManager. If better credit judgements are from default histories, then perhaps an actuarial model such as CreditRisk+ will dominate.

Similarly, the selection and application of available models is best made after first considering what data are available for user input. An analyst may deeply believe that, say, credit spreads contain the best credit-quality information. Nevertheless, if spreads for his particular

To continue reading...

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: