nnpiv.semiparametrics.DML_npiv

class nnpiv.semiparametrics.DML_npiv(Y, D, Z, W, X1=None, V=None, v_values=None, include_V=True, ci_type='pointwise', loc_kernel='gau', bw_loc='silverman', estimator='MR', model1=<nnpiv.rkhs.rkhsiv.ApproxRKHSIVCV object>, nn_1=False, modelq1=<nnpiv.rkhs.rkhsiv.ApproxRKHSIVCV object>, nn_q1=False, alpha=0.05, n_folds=5, n_rep=1, random_seed=123, prop_score=sklearn.linear_model.LogisticRegression, CHIM=False, verbose=True, fitargs1=None, fitargsq1=None, opts=None)[source]

Debiased Machine Learning for Nonparametric Instrumental Variables (DML-npiv) class.

Parameters

  • Y (array-like) – Outcome variable.

  • D (array-like) – Treatment variable.

  • Z (array-like) – Instrumental variable.

  • W (array-like) – Negative control outcome.

  • X1 (array-like or None) – Additional covariates.

  • V (array-like or None) – Localization covariates. Supplying V changes the estimand to a finite-bandwidth kernel-ratio target.

  • v_values (array-like or None) – Evaluation points for localization.

  • include_V (bool) – Whether to include localization covariates in the model.

  • ci_type (str) – Type of confidence interval (‘pointwise’, ‘uniform’).

  • loc_kernel (str) – Kernel for localization. Options include ‘gau’, ‘epa’, ‘uni’, and ‘tri’.

  • bw_loc (str) – Bandwidth for localization.

  • estimator (str) – Estimator type (‘MR’, ‘OR’, ‘IPW’).

  • model1 (estimator) – Model for the first stage.

  • nn_1 (bool) – Whether to use a neural network for the first stage.

  • modelq1 (estimator) – Model for the second stage.

  • nn_q1 (bool) – Whether to use a neural network for the second stage.

  • alpha (float) – Significance level for confidence intervals.

  • n_folds (int) – Number of folds for estimation.

  • n_rep (int) – Number of repetitions for estimation.

  • random_seed (int) – Seed for the random number generator.

  • prop_score (estimator) – Model for the propensity score.

  • CHIM (bool) – Whether to drop observations with extreme propensity scores using CHIM (2009).

  • verbose (bool) – Whether to print progress information.

  • fitargs1 (dict or None) – Arguments for fitting the first-stage model.

  • fitargsq1 (dict or None) – Arguments for fitting the second-stage model.

  • opts (dict or None) – Additional options.