Abstract:
The Prudential Regulation has raised the issue of estimation errors due to Internal Rating Based (IRB) estimation process that may produce underestimation of the risk measures. In the context of credit risk, lower bounds (i.e. floors) for the estimated parameters are introduced to limit the impact of such possible underestimation. These floors are heuristically justified by the difficulties to estimate the parameters when the default event becomes rare, as in the case of Low Default Probability Portfolios (LDP). In this paper, on the basis of a standard Asymptotic Single Risk Factor (ASRF) model, and by means of Monte Carlo simulations, we provide a robust justification to PD floors, and a framework for their calibration. Our results give hints that the introduction of a floor can indeed mitigate the possibility that the risk measures become less reliable.
