How do you check stress test scenarios and assumptions?

1 Define your purpose

Before you design or run a stress test, you need to have a clear idea of what you want to achieve with it. Do you want to evaluate the impact of a specific event, such as a market crash, a cyberattack, or a pandemic? Do you want to test the robustness of your strategy, policies, or controls under various conditions? Do you want to comply with regulatory requirements, industry standards, or best practices? Your purpose will guide your choice of scenarios and assumptions, as well as the metrics and outputs you will use to measure and communicate the results.

2 Select realistic and relevant scenarios

A scenario is a hypothetical situation that describes the sources, drivers, and outcomes of a stress event. You can use historical, hypothetical, or hybrid scenarios, depending on your purpose and data availability. However, you should always check that your scenarios are realistic and relevant for your system, portfolio, or business. Realistic scenarios are plausible and consistent with historical or expected patterns and trends. Relevant scenarios are tailored to your specific context and objectives, and cover the most significant risks and uncertainties that you face.

3 Choose appropriate and consistent assumptions

An assumption is a simplification or approximation that you make to model a scenario and its effects. You can use qualitative or quantitative assumptions, depending on the level of detail and complexity you need. However, you should always check that your assumptions are appropriate and consistent for your stress test. Appropriate assumptions are based on sound logic, evidence, and expert judgment. They are also transparent and documented, so that you can explain and justify them. Consistent assumptions are aligned with your purpose, scenarios, and data. They are also coherent and compatible with each other, and avoid contradictions or overlaps.

4 Validate your data and models

Data and models are the inputs and tools that you use to run a stress test and generate the outputs. You can use different types of data and models, depending on your purpose, scenarios, and assumptions. However, you should always check that your data and models are valid and reliable for your stress test. Valid data and models are accurate, complete, and up-to-date. They are also relevant and representative of your system, portfolio, or business. Reliable data and models are consistent, robust, and scalable. They are also tested, verified, and calibrated, so that you can trust their quality and performance.

5 Compare and benchmark your results

Results are the outputs and outcomes that you obtain from a stress test. You can use different types of results, depending on your purpose, scenarios, assumptions, data, and models. However, you should always check that your results are meaningful and useful for your stress test. Meaningful results are clear, concise, and comprehensive. They are also actionable and insightful, as they highlight the key findings, implications, and recommendations from your stress test. Useful results are comparable and benchmarked against other sources, such as historical data, peer data, industry standards, or regulatory thresholds. They are also communicated and reported effectively, so that you can share and discuss them with your stakeholders.

6 Review and update your stress test

A stress test is not a static or one-off exercise. It is a dynamic and iterative process that requires regular review and update. You should always check that your stress test is current and relevant for your system, portfolio, or business. Current stress test reflects the latest data, scenarios, assumptions, models, and results. It also incorporates the feedback, lessons learned, and best practices from your previous stress tests. Relevant stress test adapts to the changing environment, risks, and opportunities that you face. It also aligns with your strategic goals, operational needs, and regulatory expectations.

7 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?