01 / THE PRACTICAL QUESTION
A decision, before a model.
A risk committee asks which combination of market shocks could produce a loss large enough to breach its tolerance.
Find shocks that reach a specified loss, then assess whether those assumptions are plausible.
02 / DATA & COMPARISON
The idea in plain language.
Conventional stress testing starts with shocks and calculates losses. Reverse stress testing starts with the loss and searches for shocks that cause it.
- Transmission model
- The mapping from a change in a risk factor to a change in portfolio value.
Data. Constructed scenarios; official scenarios are contextual references.
Baseline. Unstressed exposure values and transparent factor shocks.
03 / THE EXPERIMENT
What the saved experiment shows.
The saved linear example reaches a 10% loss with an equity shock of −10% and a spread-factor shock of +5%, satisfying its constraint to numerical precision. The result is conditional on this transmission model.
Inspect the supporting resultEvidence record: research-validation.json#numerical_checks/07
Explore the related lab04 / RESULTS & LIMITATIONS
Evidence with its boundaries attached.
The related lab is a cross-project demonstration. Read this investigation’s evidence and limits before transferring its conclusions.
Inspect numerical checks and validation records
Loading validation evidence…
A scenario is conditional, not a probability forecast or regulatory submission.
05 / REPRODUCE
Reproduce and challenge the result.
Code, configuration, and reproduction
The project contains its implementation, configuration, tests, and walkthrough. Download the lab configuration to record the exact parameters used in an interactive run.
Project code and walkthrough