Content area

Abstract

To perform risk and portfolio management, the authors must represent the distribution of the risk factors that affect the market. The most flexible approach is in terms of scenarios and their probabilities, which includes historical scenarios, pure Monte Carlo and importance sampling. Here, they present a simple method to generate scenarios from elliptical distributions with given sample means and covariances. This is very important in applications such as mean-variance portfolio optimization, which are heavily affected by incorrect representations of the first two moments. Risk managers can now proceed to stress test the correlation C using the Cholesky decomposition of the new stress-test matrix and the J x 2I panel Y of uncorrelated standard normal simulations in the above process. Then they can analyze the impact of the stress test on a risk report, confident that the stress-test assumptions will be faithfully reflected in the simulations.

Details

Title
Simulations with exact means and covariances
Author
Meucci, Attilio
Pages
89-91
Section
CUTTING EDGE. BRIEF COMMUNICATION
Publication year
2009
Publication date
Jul 2009
Publisher
Incisive Media Limited
ISSN
09528776
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
201309990
Copyright
Copyright Incisive Media Plc Jul 2009