DifferenceInDifferences#

class causalpy.experiments.diff_in_diff.DifferenceInDifferences[source]#

A class to analyse data from Difference in Difference settings.

Note

There is no pre/post intervention data distinction for DiD, we fit all the data available.

Parameters:

  • data (NativeDataFrame) – Any eager dataframe Narwhals supports, such as pandas, Polars, or PyArrow. Converted to pandas internally.

  • formula (str) – A statistical model formula.

  • time_variable_name (str) – Name of the data column for the time variable.

  • group_variable_name (str) – Name of the data column for the group variable.

  • post_treatment_variable_name (str) – Name of the data column indicating post-treatment period. Defaults to “post_treatment”.

  • model (PyMCModel | RegressorMixin | None) – A PyMC model for difference in differences. Defaults to LinearRegression.

Notes

Estimate extraction

Both Bayesian and OLS backends store the fitted group-by-post interaction coefficient as causal_impact. The class also constructs treated-post counterfactual predictions for visualization by setting that interaction term to zero, but those predictions do not determine the reported scalar effect. In an additive identity-link model, the coefficient and the corresponding predicted contrast are algebraically identical.

Examples

>>> import causalpy as cp
>>> df = cp.load_data("did")
>>> seed = 42
>>> result = cp.DifferenceInDifferences(
...     df,
...     formula="y ~ 1 + group*post_treatment",
...     time_variable_name="t",
...     group_variable_name="group",
...     model=cp.pymc_models.LinearRegression(
...         sample_kwargs={
...             "target_accept": 0.95,
...             "random_seed": seed,
...             "progressbar": False,
...         }
...     ),
... )

Methods

DifferenceInDifferences.algorithm()

Run the experiment algorithm: fit model, predict, and calculate causal impact.

DifferenceInDifferences.effect_summary(*[, ...])

Generate a decision-ready summary of causal effects for Difference-in-Differences.

DifferenceInDifferences.generate_report(*[, ...])

Generate a self-contained HTML report for this experiment.

DifferenceInDifferences.input_validation()

Validate the input data and model formula for correctness.

DifferenceInDifferences.plot(*[, round_to, ...])

Plot the difference-in-differences results.

DifferenceInDifferences.print_coefficients([...])

Ask the model to print its coefficients.

DifferenceInDifferences.set_maketables_options(*)

Set optional maketables rendering options for this experiment.

DifferenceInDifferences.summary([round_to])

Print summary of main results and model coefficients.

Attributes

idata

Return fitted DataTree when the model backend supports it.

supports_bayes

supports_ols

supports_pymc_forecast

labels

data

__init__(data, formula, time_variable_name, group_variable_name, post_treatment_variable_name='post_treatment', model=None)[source]#
Parameters:
Return type:

None

classmethod __new__(*args, **kwargs)#