graded_intervention_its#
Graded Intervention Time Series Experiment
This module implements experiments for estimating the causal effects of graded interventions (e.g., media spend, policy intensity) in single-market time series using transfer functions that model saturation and adstock (carryover) effects.
The experiment works with the TransferFunctionOLS model class (from skl_models) to provide a complete causal inference workflow including visualization, diagnostics, and counterfactual effect estimation.
Classes
Interrupted time series experiment with graded interventions and transfer functions. |