Scenario help

Follow the scenario workflow, then review the concepts and variation types used to create and compare forecast scenarios.

Browse help topics and this page

Help topics

Compare scenarios step by step

Choose a baseline run
Start with a completed run that contains at least one known future variable. This run remains the fixed baseline for the project.
Create a project and alternatives
Create a project from the baseline, then add alternatives by changing one or more scenario conditions. Open an alternative to review it, rename it, or use Reuse configuration to create another alternative from its settings. Reuse removes fields the current API no longer supports and tells you exactly what was omitted.
Use JSON configuration when needed
The JSON creation option can load a scenario_parameters template from the current Forecast API contract. Rejected keys must be removed before submission; keys the API would silently ignore are identified but do not block submission. Existing JSON entry and backend validation remain available if the contract cannot be loaded.
Select and compare scenarios
Select one or more alternatives and choose Compare selected. The baseline is included automatically, so it does not need a selectable row.
Review or change a comparison
Comparison results open in the Comparison view. Choose Change scenarios to return to the project with the current selection, where you can select different alternatives or leave the comparison workflow.

Term	Meaning	Details
Scenario	A simulation of forecast changes caused by numeric or categorical changes to known variables during the prediction horizon.	Corresponds to a Prediction pipeline run.
Baseline Scenario	The fixed reference scenario generated from the project's baseline run.	Corresponds to a Complete pipeline run configured with at least one known variable.
Project	A collection of scenarios that share the same baseline scenario.	The baseline is included automatically whenever alternatives are compared.

Lever type	Variation	Meaning	Example
Numeric Changes to numeric future-known exogenous features, such as testing how forecasts change when price increases by 5%.	Absolute	Set the variable to a fixed value.	Passing k changes the existing value V to k.
Numeric Changes to numeric future-known exogenous features, such as testing how forecasts change when price increases by 5%.	Relative	Modify the existing value by a proportional change.	Passing k changes the existing value V to V × (1 + k).
Categorical Changes to categorical future-known exogenous features, such as testing different discounts for loyalty groups.	Lock	Set the variable to a fixed level.	Passing p changes every existing level to p.
Categorical Changes to categorical future-known exogenous features, such as testing different discounts for loyalty groups.	Switch	Map existing levels to new levels.	Passing {p1: p2, p3: p4} changes p1 to p2 and p3 to p4.