experiment

CausalPy experiment bridge for MMM lift test calibration.

This module provides an integration layer between CausalPy causal inference experiments and the MMM lift test calibration workflow. It allows users to run quasi-experiments (e.g. Interrupted Time Series, Synthetic Control, Difference-in-Differences) and convert results into the DataFrame format expected by MMM.add_lift_test_measurements().

Examples

Run a Synthetic Control experiment and convert to a lift test:

from pymc_marketing.mmm.experiment import run_experiment

result = run_experiment(
    experiment_type="sc",
    data=df,
    treatment_time=70,
    formula="actual ~ 0 + a + b + c + d + e + f + g",
)

df_lift = result.to_lift_test(
    channel="tv",
    x=1000.0,
    delta_x=200.0,
    geo="US",
)

mmm.add_lift_test_measurements(df_lift)

Functions

run_experiment(experiment_type, data, **kwargs)

Run a CausalPy experiment and return a wrapped result.

Classes

ExperimentResult(result, experiment_type)	Wrapper around a CausalPy experiment result.
ExperimentType(value)	Supported CausalPy experiment types.