class econml.inference.BootstrapInference(n_bootstrap_samples=100, n_jobs=- 1, bootstrap_type='pivot', verbose=0)[source]

Bases: econml.inference._inference.Inference

Inference instance to perform bootstrapping.

This class can be used for inference with any CATE estimator.

  • n_bootstrap_samples (int, optional (default 100)) – How many draws to perform.

  • n_jobs (int, optional (default -1)) – The maximum number of concurrently running jobs, as in joblib.Parallel.

  • verbose (int, default: 0) – Verbosity level

  • bootstrap_type (‘percentile’, ‘pivot’, or ‘normal’, default ‘pivot’) – Bootstrap method used to compute results. ‘percentile’ will result in using the empiracal CDF of the replicated computations of the statistics. ‘pivot’ will also use the replicates but create a pivot interval that also relies on the estimate over the entire dataset. ‘normal’ will instead compute a pivot interval assuming the replicates are normally distributed.

__init__(n_bootstrap_samples=100, n_jobs=- 1, bootstrap_type='pivot', verbose=0)[source]

Initialize self. See help(type(self)) for accurate signature.


__init__([n_bootstrap_samples, n_jobs, …])

Initialize self.

ate_inference([X, T0, T1])

ate_interval([X, T0, T1, alpha])


const_marginal_ate_interval([X, alpha])

fit(estimator, *args, **kwargs)

Fits the inference model.

marginal_ate_inference(T[, X])

marginal_ate_interval(T[, X, alpha])

prefit(estimator, *args, **kwargs)

Performs any necessary logic before the estimator’s fit has been called.

fit(estimator, *args, **kwargs)[source]

Fits the inference model.

This is called after the estimator’s fit.

prefit(estimator, *args, **kwargs)

Performs any necessary logic before the estimator’s fit has been called.