This paper contributes to the literature for mixture models by leveraging an efficient algorithm for computing the density function of the loss distribution and extending the model in two key areas: constructing the systemic variable from a continuous…
By: Jorge A Chan-Lau
The gate array way
Risk awards 2012
A popular copula
Top quant says a CVA model that is 80% accurate but takes 20% of the time is "very attractive"
RBS adds former Barclays Capital and Lehman quant to help build firm-wide CVA model
An analytical framework for credit portfolio risk measures
Multi-period portfolio optimisation for structured investment strategies
Name concentration correction
Several financial institutions use single-period models to determine their credit portfolio loss distribution, calculate their loss volatility and assign economic capital.
Credit portfolio models often assume that recovery rates are independent of default probabilities. Here, Jon Frye presents empirical evidence showing that such assumptions are wrong. Using US historical default data, he shows that not only are recovery…
Credit portfolio models often assume that recovery rates are independent of defaultprobabilities. Here, Jon Frye presents empirical evidence showing that such assumptions arewrong. Using US historical default data, he shows that not only are recovery…
Recovery rates - Cutting edge
Of the various analytical approaches to credit portfolio modelling, CreditRisk+ has become the most popular due to its tractability. However, the model suffers from the restrictive assumption of sector independence. Moreover, the recursion relation for…
For credit portfolios, analytical methods work best for tail risk, while Monte Carlo is used to model expected loss. However, products such as CDOs require a model for the entire distribution. Sandro Merino and Mark Nyfeler meet the challenge by…