ShortenedWindowScenario
Agents scenarios · Implementation: mlsysim/engine/scenario_definitions/agents.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
An observation window cut to window_share of its adopted length; samples scale with it.
Notes
Equation: Half-width changes by 1/sqrt(window_share) when independent sample count is proportional to window length.
Assumptions: The rate of independent samples and outcome variance stay constant; window_share is positive.
Excludes: Changes in bias, autocorrelation, or nonstationarity when the observation window is shortened.
References
Derivation: Half-width changes by 1/sqrt(window_share) when independent sample count is proportional to window length. This is declared scenario accounting rather than a fitted empirical law.
Examples
>>> from mlsysim import Scenarios
>>> scenario = Scenarios.HalvedHealthWindow
>>> value = scenario.interval_widening
>>> round(float(value), 3)
1.414Attributes
| Name | Description |
|---|---|
| interval_widening | Normal-error half-width multiplier when sample count scales by positive |
| window_share |
interval_widening
engine.scenario_types.ShortenedWindowScenario.interval_widening
Normal-error half-width multiplier when sample count scales by positive window_share.
Equation: 1 / _math.sqrt(window_share).