InversePropensityWeighting

class causalpy.experiments.inverse_propensity_weighting.InversePropensityWeighting

A class to analyse inverse propensity weighting experiments.

Parameters:

• data (DataFrame) – A pandas dataframe

• formula (str) – A statistical model formula for the propensity model

:param outcome_variable

A string denoting the outcome variable in datq to be reweighted

Parameters:

• weighting_scheme (str) – A string denoting which weighting scheme to use among: ‘raw’, ‘robust’, ‘doubly robust’ or ‘overlap’. See Aronow and Miller “Foundations of Agnostic Statistics” for discussion and computation of these weighting schemes.

• model – A PyMC model

Example

>>> import causalpy as cp
>>> df = cp.load_data("nhefs")
>>> seed = 42
>>> result = cp.InversePropensityWeighting(
...     df,
...     formula="trt ~ 1 + age + race",
...     outcome_variable ="outcome",
...     weighting_scheme="robust",
...     model=cp.pymc_models.PropensityScore(
...         sample_kwargs={
...             "draws": 100,
...             "target_accept": 0.95,
...             "random_seed": seed,
...             "progressbar": False,
...         },
...     ),
... )

Methods

InversePropensityWeighting.__init__(data, ...)
InversePropensityWeighting.bayesian_plot(...)	Abstract method for plotting the model.
InversePropensityWeighting.get_ate(i, idata)
InversePropensityWeighting.input_validation()	Validate the input data and model formula for correctness
InversePropensityWeighting.make_doubly_robust_adjustment(ps)	The doubly robust weighting scheme is also discussed in Aronow and Miller, but a bit more generally than our implementation here.
InversePropensityWeighting.make_overlap_adjustments(ps)	This weighting scheme was adapted from Lucy D’Agostino McGowan's blog on Propensity Score Weights referenced in the primary CausalPy explainer notebook
InversePropensityWeighting.make_raw_adjustments(ps)	This estimator is discussed in Aronow and Miller as the simplest of base form of inverse propensity weighting schemes
InversePropensityWeighting.make_robust_adjustments(ps)	This estimator is discussed in Aronow and Miller's book as being related to the Horvitz Thompson method
InversePropensityWeighting.ols_plot(*args, ...)	Abstract method for plotting the model.
InversePropensityWeighting.plot(*args, **kwargs)	Plot the model.
InversePropensityWeighting.plot_ate([idata, ...])
InversePropensityWeighting.plot_balance_ecdf(...)	Plotting function takes a single covariate and shows the differences in the ECDF between the treatment and control groups before and after weighting.
InversePropensityWeighting.print_coefficients([...])	Ask the model to print its coefficients.
InversePropensityWeighting.weighted_percentile(...)	perc : percentile in [0-1]!

Attributes

idata	Return the InferenceData object of the model.
supports_bayes
supports_ols

__init__(data, formula, outcome_variable, weighting_scheme, model=None, **kwargs)

Parameters:

• data (DataFrame)

• formula (str)

• outcome_variable (str)

• weighting_scheme (str)

__new__(**kwargs)