causallib.estimation.marginal_outcome module

  1. Copyright 2019 IBM Corp.

Licensed under the Apache License, Version 2.0 (the “License”); you may not use this file except in compliance with the License. You may obtain a copy of the License at

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an “AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Created on Apr 25, 2018

class causallib.estimation.marginal_outcome.MarginalOutcomeEstimator(learner, use_stabilized=False, *args, **kwargs)[source]

Bases: causallib.estimation.base_weight.WeightEstimator, causallib.estimation.base_estimator.PopulationOutcomeEstimator

A marginal outcome predictor. Assumes the sample is marginally exchangeable, and therefore does not correct (adjust, control) for covariates. Predicts the outcome/effect as if the sample came from a randomized control trial: $Pr[Y|A]$.

Parameters
compute_weight_matrix(X, a, use_stabilized=None, **kwargs)[source]

Computes individual weight across all possible treatment values. f(Pr[A=a_j | X_i]) for all individual i and treatment j.

Parameters

  • X (pd.DataFrame) – Covariate matrix of size (num_subjects, num_features).

  • a (pd.Series) – Treatment assignment of size (num_subjects,).

  • use_stabilized (bool) – Whether to re-weigh the learned weights with the prevalence of the treatment. This overrides the use_stabilized parameter provided at initialization. See Also: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4351790/#S6title

  • **kwargs –

Returns

A matrix of size (num_subjects, num_treatments) with weight for every individual and every

treatment.

Return type

pd.DataFrame

compute_weights(X, a, treatment_values=None, use_stabilized=None, **kwargs)[source]

Computes individual weight given the individual’s treatment assignment. f(Pr[A=a_i | X_i]) for each individual i.

Parameters

  • X (pd.DataFrame) – Covariate matrix of size (num_subjects, num_features).

  • a (pd.Series) – Treatment assignment of size (num_subjects,).

  • treatment_values (Any | None) – A desired value/s to extract weights to (i.e. weights to what treatment value should be calculated). If not specified, then the weights are chosen by the individual’s actual treatment assignment.

  • use_stabilized (bool) – Whether to re-weigh the learned weights with the prevalence of the treatment. This overrides the use_stabilized parameter provided at initialization. See Also: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4351790/#S6title

  • **kwargs –

Returns

A vector of size (num_subjects,) with a weight for each individual

Return type

pd.Series

estimate_population_outcome(X, a, y, w=None, treatment_values=None)[source]

Calculates potential population outcome for each treatment value.

Parameters

  • X (pd.DataFrame) – Covariate matrix of size (num_subjects, num_features).

  • a (pd.Series) – Treatment assignment of size (num_subjects,).

  • y (pd.Series) – Observed outcome of size (num_subjects,).

  • w (pd.Series | None) – Individual (sample) weights calculated. Used to achieved unbiased average outcome. If not provided, will be calculated on the data.

  • treatment_values (Any) – Desired treatment value/s to stratify upon before aggregating individual into population outcome. If not supplied, calculates for all available treatment values.

Returns

Series which index are treatment values, and the values are numbers - the

aggregated outcome for the strata of people whose assigned treatment is the key.

Return type

pd.Series[Any, float]

fit(X=None, a=None, y=None)[source]

Dummy implementation to match the API. MarginalOutcomeEstimator acts as a WeightEstimator that weights each sample as 1

Parameters

  • X (pd.DataFrame) – Covariate matrix of size (num_subjects, num_features).

  • a (pd.Series) – Treatment assignment of size (num_subjects,).

  • y (pd.Series) – Observed outcome of size (num_subjects,).

Returns

a fitted model.

Return type

MarginalOutcomeEstimator