| Units: | - |
| Default Value: | 1E-06 |
| Validation Rule: | ≥0 |
| Description: | Minimum sample weight for rolling horizon iterations |
Stochastic Minimum Sample Weight applies to the Rolling Horizon algorithm for solving multi-stage stochastic optimization - see MT Schedule Stochastic Algorithm. It is a tolerance used to determine how many stages should be included in one rolling step in rolling horizon. Generally, at each node in the Scenario tree, the branches (Full or Hanging) are assumed to have uniform weights. The last stage in one rolling step is the one which has probabilities(weights) below this tolerance, no more hanging branches are created after this stage.
The number of branching stages calculated based on this parameter can be overridden with the Rolling Stage Increment setting.