Core API Reference¶
Generic DualBounds¶
|
Computes dual bounds on \(E[f(Y(0),Y(1), X)].\) |
|
Computes generalized dual bounds via the delta method. |
|
Computes plug-in bounds on \(E[f(Y(0),Y(1))]\) without using covariates. |
Bootstrap methods¶
|
Combines evidence across multiple DualBounds classes using the multiplier bootstrap. |
|
Computes multiplier bootstrap lower confidence bounds. |
Lee bounds¶
|
Computes dual bounds on the ATE under selection bias. |
|
Computes plug-in Lee bounds without using covariates. |
|
Helper function to compute semi-analytical Lee Bounds. |
Variance bounds¶
|
Computes dual bounds on \(Var(Y(1) - Y(0)).\) |
|
Computes lower bounds on \(Var(E[Y(1) - Y(0) | X]).\) |
|
Improved lower bounds on \(Var(E[Y(1) - Y(0) | X])\). |
IV Bounds (in beta)¶
|
Beta version. |
Distributional regression¶
|
A generic class for distributional regressions, meant for subclassing. |
|
Binary regression which inherits from |
|
Distributional regression for continuous outcomes. |
|
A continuous distributional regression based on quantile regression. |
|
A logistic regression solver which ensures that beta[0] >= 0. |
Class which selects a distributional regression model. |
|
|
Performs cross-fitting for a model inheriting from |
Utility functions¶
Synthetic data generation¶
|
Samples a synthetic regression dataset. |
|
Generates synthetic datasets with selection bias. |
Interpolation¶
|
Adaptively chooses between linear and nearest-neighbor interpolation. |
|
Nearest-neighbor interpolation. |
|
Linear interpolation. |
Miscellaneous¶
|
Batched discrete (categorical) distribution. |
|
Helper to computes confidence intervals. |
|
Very close to numpy.percentile, but supports weights. |
|
Adjust categorical distribution to have support of size |