Underdetermination and variability of the results in macro-to-micro stress tests: A machine learning approach

Alexander Denev and Orazio Angelini

Journal of Risk Management in Financial Institutions, 2017, vol. 10, issue 2, 130-149

Abstract: We investigate the impact of the uncertainties surrounding the modelling process when conducting a stress test. These uncertainties are due to several choices left to the modeller with regards to, among others, the variables to select, the data samples used for the calibration of the different models and how these models are combined together. We run tests to quantify the impact of these sources of uncertainty by using as an example the Federal Reserve System’s Comprehensive Capital Analysis and Review (FED CCAR) 2016 scenario. We conclude that the impact could be non-negligible as it adds substantial variability to the final results. We employ Probabilistic Graphical Models — a machine learning technique — to corroborate our findings.

Keywords: stress testing; scenario analysis; probabilistic graphical models; visualisation; model risk; machine learning (search for similar items in EconPapers)
JEL-codes: E5 G2 (search for similar items in EconPapers)
Date: 2017
References: Add references at CitEc
Citations:

Downloads: (external link)
https://hstalks.com/article/1078/download/ (application/pdf)
https://hstalks.com/article/1078/ (text/html)
Requires a paid subscription for full access.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:aza:rmfi00:y:2017:v:10:i:2:p:130-149

Access Statistics for this article

More articles in Journal of Risk Management in Financial Institutions from Henry Stewart Publications
Bibliographic data for series maintained by Henry Stewart Talks ().