Bank Stress Testing: Regulatory, Historical, and Reverse Scenarios
Summary
The document describes several ways banks and financial institutions test portfolio and balance-sheet resilience. These include regulator-provided scenarios, shocks based on past events, and reverse stress analysis, which searches for plausible conditions that would cause the greatest harm. It notes that Monte Carlo scenarios generated for value-at-risk calculations can also reveal damaging tail outcomes for review and hedging.
Machine learning is not required for reverse stress testing, and the answer does not identify a general ML method for generating scenarios. It mentions AI or machine learning as part of some work on synthetic stress scenarios, but gives no implementation details or evidence about effectiveness. Neural networks and algorithmic differentiation are cited separately as ways to speed up complex product valuation or sensitivity calculations, rather than as established methods for creating stressors. The examples are illustrative and do not establish how widely any technique is used.
Key ideas
- Banks may use common regulatory scenarios alongside shocks based on historical events.
- Reverse stress testing searches for plausible scenarios that would most damage a portfolio or balance sheet.
- Monte Carlo tail scenarios from value-at-risk calculations can help identify risks and possible hedges.
- Machine learning is not necessary for scenario analysis, and its use in scenario generation is presented as limited and unspecified.
- Neural networks and algorithmic differentiation can accelerate valuation and sensitivity calculations.
Tags
Full text
# Stress testing by Banks # Stress testing by Banks AFAIK typically banks `stress test` it trading portfolio by assuming stressed value of risk factors or by considering times series corresponding to some historical stress period e.g. 2008-2009 financial crisis period. So my question is, is there any application of advanced `machine learning techniques` used by bank in their stress testing application? ## Answer by Dimitri Vulis (score 3) https://quant.stackexchange.com/a/66631 U.S. centric answer. Banks/financial institutions are given standard stress scenarios by regulators for CCAR and DFAST (Dodd-Frank Act Stress Testing). It's a good bet that many institutions in various jurisdictions will also be given stress scenarios for climate risk in the next few years. These are the same scenarios for many institutions. Also many institutions indeed run stress tests based on some historical events, for example, Lehman bankruptcy in 2008, or Russian sovereign default in 1998, or the COVID-19 lockdown in March 2020. However many institutions also do a "reverse" analysis. They search for plausible stress scenarios that would hurt them the most. It is not necessary to use AI/ML for this. For example, if the institution has to calculate VaR, and chooses to use Monte Carlo to generate scenarios for VaR, then you get "for free" the MC scenarios in the tail of the distribution that would hurt you. You can analyse these scenarios and think of ways to hedge them. However the scenarios from MC are not exhaustive. It is possible to have scenarios that don't look like anything like recent history or anything from a Monte Carlo simulation that looks like recent history. One place I know to be doing interesting work on generating stress scenarios that would hurt a given portfolio / balance sheet (sometimes using AI/ML) is Straterix / Alla Gil. I think some of her recent GARP posts are relevant to your question: Risk Management in a Sea of Unknown Unknowns: The Complex Quest for Resilience How to Mine Synthetic Data: Pros and Cons of a Shiny New Tool for Risk Managers How to Stress Test for Extremely Unexpected Scenarios What’s Missing in Asset and Liability Management? ## Answer by Bob Jansen (score 1) https://quant.stackexchange.com/a/66629 Two applications of machine learing (related) techniques are in valuation using neural nets of complex products, e.g. "Deep xVA solver -- A neural network based counterparty credit risk management framework" or using Adjoint Algorithmic Differentiation, e.g. "Fast Greeks by algorithmic differentiation". Both of these techniques aim to speed up the computation. I'm not aware of using ML to come up with stress scenarios or stressors.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.