Publication at Faculty of Social Sciences |
2012
Abstract
This paper describes the stress-testing framework used in the Czech central bank and focuses on the general question of how to calibrate the models and parameters used to stress test the most important risks in the banking system. The paper argues that stress tests should be calibrated conservatively to overestimate the risks so that sufficient buffers are in place for when adverse shocks materialize.
However, to ensure that the stress test framework is conservative enough over time, backtesting, i.e., comparison of the actual values of key financial variables with the predictions generated by the stress-testing models, should be a standard part of the stress-testing framework.