Calibrating the CreditRisk+ Model at Different Time Scales and in Presence of Temporal Autocorrelation †

The CreditRisk+ model is one of the industry standards for valuation default risk in credit loans portfolios. calibration requires, inter alia, specification parameters describing structure dependence among events. This work addresses these parameters. In particular, we study procedure on sampling period rate time series, that might be different from horizon onto which used forecasting, as it often case real life applications. autocorrelated series and role statistical error a function are also discussed. findings proposed technique illustrated with support an application to data.