We propose a fast algorithm for computing the expected tranche loss in the Gaussian factor model with arbitrary accuracy using Hermite expansions. No assumptions about homogeneity of the portfolio are made. The algorithm is a generalization of the algorithm proposed in \cite{PO}. The advantage...

We propose a simple approach to combining internal and external loss data in the case when internal and external data come from the same distribution. We assume that the internal data is uncensored but the external data includes only losses above a known threshold. This approach is an...