Dependent default and recovery: Markov chain Monte Carlo study of downturn Loss Given Default credit risk model

Authors

  • Pavel V Shevchenko CSIRO Mathematics, Informatics and Statistics
  • Xiaolin Luo CSIRO Mathematics, Informatics and Statistics

DOI:

https://doi.org/10.21914/anziamj.v53i0.5080

Keywords:

credit risk, Markov chain Monte Carlo, Bayesian inference

Abstract

There is empirical evidence that recovery rates tend to go down just when the number of defaults goes up in economic downturns. This has to be taken into account in estimation of the capital against credit risk required by Basel II to cover losses during the adverse economic downturns; the so-called ``downturn Loss Given Default" requirement. This article presents a methodology for estimation of the Loss Given Default credit risk model with the default and recovery dependent via the latent systematic risk factor using a Bayesian inference approach and Markov chain Monte Carlo method. This approach allows joint estimation of all model parameters and latent systematic factor, and all relevant uncertainties. For illustration, we fit the model using Moody's annual default and recovery rates for corporate bonds for the period 1982--2010. References

Published

2012-06-06

Issue

Vol. 53 (2011)

Section

Proceedings Engineering Mathematics and Applications Conference