Stuart M. Turnbull
Bauer Chaired Professor, Bauer College of Business, University of Houston
In this issue we have four papers. The first, by Leif Andersen, examines the structure and limitations of different types of factor models used for pricing credit portfolio derivatives, given a term structure of prices. An unresolved issue is the trade-off between model complexity and data and calibration limitations. The remaining three papers are of a more applied nature.
The second paper, by Paul Glasserman and Jesus Ruiz-Mata, asks what is the best way to simulate the distribution of losses from defaults in a credit portfolio. They reach a surprising conclusion! The paper by Kevin Thompson and Alistair McLeod takes a top-down approach to loss attribution, by estimating the expected loss contribution of individual obligors for a given portfolio loss. The final paper, by Alfred Hamerle and Daniel Rösch, uses maximum likelihood to estimate the parameters of three common credit risk models. They compare the forecast loss distributions for the three models and argue that the results are similar.
Analytic calculation of conditional default statistics and risk contributions using the Ensemble method