E-17 — Scenario
A defined stress or hypothetical scenario — a named set of market shocks and assumptions applied across the portfolio to ask “what would happen if.” The definition a stress test or scenario analysis runs against.
Purpose
Scenario analysis and stress testing ask what the portfolio would be worth, and what risk it would carry, under conditions other than today’s. A Scenario is the definition of one such condition: a named, governed set of shocks — an equity-market fall, a rates move, a credit-spread widening, an FX dislocation, a liquidity freeze — that can be applied consistently across every holding. Modelling the scenario as an entity, rather than as a transient analyst calculation, is what makes a stress result reproducible, auditable and comparable over time: the same scenario run this quarter and last is genuinely the same scenario.
A Scenario is a definition; running it produces Risk Measurements (E-19) tagged to it.
Attribute schema
| Column | Type | Definition |
|---|---|---|
scenario_id | varchar | Primary key. |
scenario_name | varchar | Canonical name. |
scenario_type | varchar | historical (a replay of a past episode) / hypothetical (a forward-constructed condition) / reverse (a target loss, solved back to the conditions that produce it) / regulatory (a prescribed supervisory scenario). |
description | varchar | What the scenario represents and why it is run. |
shock_set | document (JSON) | The structured set of shocks — the factors moved and by how much. A typed specification the stress engine interprets, in the same computation-as-data spirit as PM-10 Fund Terms. |
horizon | varchar | The time horizon the scenario is expressed over. |
effective_from | date | When this version of the scenario became active. |
effective_to | date | When it was retired or superseded; null while active. |
owner | varchar | The risk owner accountable for the scenario definition. |
Notes
- Scenarios are versioned, not edited in place. A scenario whose shocks are recalibrated is a new version; the prior version is retained, so a historical stress result remains traceable to the exact scenario definition it was produced from.
- The
shock_setis structured data, not prose — the stress engine reads and applies it, the same discipline as theformula_specof a computed fund term. - A Scenario applies across asset classes — an equity shock, a rates move and a spread widening hit a multi-asset portfolio together — which is why it is a core entity, not specific to any one specialisation pack.
Out of scope
- The result of running a scenario — that is E-19 Risk Measurement, tagged to the scenario through
scenario_id; E-17 is the definition, E-19 is the produced number. - The configured risk constraint a stress result is judged against — that is E-16 Risk Limit; E-17 is the hypothetical condition, not the threshold.
- The
shock_setdocument grammar — the typed vocabulary the shocks are expressed in — named as an open extension; the entity carries the shock set as structured data without the formal grammar.
Owned and consumed by
- Owned by: SD-07.6 Scenario Analysis & Stress Testing.
- Consumed by: SD-07.1 Market Risk Management, SD-07.3 Liquidity Risk Management, SD-07.7 Investment Risk Reporting & Limits Governance; produces Risk Measurements (E-19).
Open extensions
- The
shock_setdocument grammar — the typed vocabulary a scenario’s shocks are expressed in. - Scenario libraries and the relationship between a reverse-stress scenario and the loss target that defines it.