Stressed in Monte Carlo

Stress tests are playing a bigger role in the operation of financial institutions, both from a risk management and a regulatory perspective. But Monte Carlo can behave badly when fed extreme parameters and give misleading results. Here, Christian Fries shows how to adapt the method to account for boundary conditions, and thereby render it more stable under stressed scenarios

A stress test is an important tool for assessing risk in a portfolio. In this article, we consider a stress test implemented by an evaluation under stressed model parameters. These could stem from a calibration to stressed market data created by a historical simulation for value-at-risk (or some other risk measure), for instance.

Click on the link below to read the full version of this article.

Stressed in Monte Carlo