[](https://dev.azure.com/ms/EconML/_build/latest?definitionId=49&branchName=master)
[](https://pypi.org/project/econml/)
[](https://pypi.org/project/econml/)
[](https://pypi.org/project/econml/)
<h1><img src="https://www.microsoft.com/en-us/research/wp-content/uploads/2016/12/MSR-ALICE-HeaderGraphic-1920x720_1-800x550.jpg" width="130px" align="left" style="margin-right: 10px;"> EconML: A Python Package for ML-Based Heterogeneous Treatment Effects Estimation</h1>
**EconML** is a Python package for estimating heterogeneous treatment effects from observational data via machine learning. This package was designed and built as part of the [ALICE project](https://www.microsoft.com/en-us/research/project/alice/) at Microsoft Research with the goal to combine state-of-the-art machine learning
techniques with econometrics to bring automation to complex causal inference problems. The promise of EconML:
* Implement recent techniques in the literature at the intersection of econometrics and machine learning
* Maintain flexibility in modeling the effect heterogeneity (via techniques such as random forests, boosting, lasso and neural nets), while preserving the causal interpretation of the learned model and often offering valid confidence intervals
* Use a unified API
* Build on standard Python packages for Machine Learning and Data Analysis
One of the biggest promises of machine learning is to automate decision making in a multitude of domains. At the core of many data-driven personalized decision scenarios is the estimation of heterogeneous treatment effects: what is the causal effect of an intervention on an outcome of interest for a sample with a particular set of features? In a nutshell, this toolkit is designed to measure the causal effect of some treatment variable(s) `T` on an outcome
variable `Y`, controlling for a set of features `X, W` and how does that effect vary as a function of `X`. The methods implemented are applicable even with observational (non-experimental or historical) datasets. For the estimation results to have a causal interpretation, some methods assume no unobserved confounders (i.e. there is no unobserved variable not included in `X, W` that simultaneously has an effect on both `T` and `Y`), while others assume access to an instrument `Z` (i.e. an observed variable `Z` that has an effect on the treatment `T` but no direct effect on the outcome `Y`). Most methods provide confidence intervals and inference results.
For detailed information about the package, consult the documentation at https://econml.azurewebsites.net/.
For information on use cases and background material on causal inference and heterogeneous treatment effects see our webpage at https://www.microsoft.com/en-us/research/project/econml/
<details>
<summary><strong><em>Table of Contents</em></strong></summary>
- [News](#news)
- [Getting Started](#getting-started)
- [Installation](#installation)
- [Usage Examples](#usage-examples)
- [Estimation Methods](#estimation-methods)
- [Interpretability](#interpretability)
- [Causal Model Selection and Cross-Validation](#causal-model-selection-and-cross-validation)
- [Inference](#inference)
- [For Developers](#for-developers)
- [Running the tests](#running-the-tests)
- [Generating the documentation](#generating-the-documentation)
- [Blogs and Publications](#blogs-and-publications)
- [Citation](#citation)
- [Contributing and Feedback](#contributing-and-feedback)
- [References](#references)
</details>
# News
**January 20, 2021:** Release v0.9.0b1, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.9.0b1)
<details><summary>Previous releases</summary>
**November 20, 2020:** Release v0.8.1, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.8.1)
**November 18, 2020:** Release v0.8.0, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.8.0)
**September 4, 2020:** Release v0.8.0b1, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.8.0b1)
**March 6, 2020:** Release v0.7.0, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.7.0)
**February 18, 2020:** Release v0.7.0b1, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.7.0b1)
**January 10, 2020:** Release v0.6.1, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.6.1)
**December 6, 2019:** Release v0.6, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.6)
**November 21, 2019:** Release v0.5, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.5).
**June 3, 2019:** Release v0.4, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.4).
**May 3, 2019:** Release v0.3, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.3).
**April 10, 2019:** Release v0.2, see release notes [here](https://github.com/Microsoft/EconML/releases/tag/v0.2).
**March 6, 2019:** Release v0.1, welcome to have a try and provide feedback.
</details>
# Getting Started
## Installation
Install the latest release from [PyPI](https://pypi.org/project/econml/):
```
pip install econml
```
To install from source, see [For Developers](#for-developers) section below.
## Usage Examples
### Estimation Methods
<details>
<summary>Double Machine Learning (aka RLearner) (click to expand)</summary>
* Linear final stage
```Python
from econml.dml import LinearDML
from sklearn.linear_model import LassoCV
from econml.inference import BootstrapInference
est = LinearDML(model_y=LassoCV(), model_t=LassoCV())
### Estimate with OLS confidence intervals
est.fit(Y, T, X=X, W=W) # W -> high-dimensional confounders, X -> features
treatment_effects = est.effect(X_test)
lb, ub = est.effect_interval(X_test, alpha=0.05) # OLS confidence intervals
### Estimate with bootstrap confidence intervals
est.fit(Y, T, X=X, W=W, inference='bootstrap') # with default bootstrap parameters
est.fit(Y, T, X=X, W=W, inference=BootstrapInference(n_bootstrap_samples=100)) # or customized
lb, ub = est.effect_interval(X_test, alpha=0.05) # Bootstrap confidence intervals
```
* Sparse linear final stage
```Python
from econml.dml import SparseLinearDML
from sklearn.linear_model import LassoCV
est = SparseLinearDML(model_y=LassoCV(), model_t=LassoCV())
est.fit(Y, T, X=X, W=W) # X -> high dimensional features
treatment_effects = est.effect(X_test)
lb, ub = est.effect_interval(X_test, alpha=0.05) # Confidence intervals via debiased lasso
```
* Generic Machine Learning last stage
```Python
from econml.dml import NonParamDML
from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier
est = NonParamDML(model_y=RandomForestRegressor(),
model_t=RandomForestClassifier(),
model_final=RandomForestRegressor(),
discrete_treatment=True)
est.fit(Y, T, X=X, W=W)
treatment_effects = est.effect(X_test)
```
</details>
<details>
<summary>Causal Forests (click to expand)</summary>
```Python
from econml.dml import CausalForestDML
from sklearn.linear_model import LassoCV
# Use defaults
est = CausalForestDML()
# Or specify hyperparameters
est = CausalForestDML(criterion='het', n_estimators=500,
min_samples_leaf=10,
max_depth=10, max_samples=0.5,
discrete_treatment=False,
model_t=LassoCV(), model_y=LassoCV())
est.fit(Y, T, X=X, W=W)
treatment_effects = est.effect(X_test)
# Confidence intervals via
EconML：ALICE（因果关系和经济学的自动学习和情报）是一项Microsoft研究项目，旨在将人工智能概念应用于经济决策。...
需积分: 44 177 浏览量
2021-02-03
06:40:10
上传
评论 1
收藏 17.34MB ZIP 举报
EconML：ALICE（因果关系和经济学的自动学习和情报）是一项Microsoft研究项目，旨在将人工智能概念应用于经济决策。 它的目标之一是构建一个工具包，将最先进的机器学习技术与计量经济学相结合，以使自动化解决复杂的因果推理问题。 迄今为止，ALICE Python SDK（econml）实现了正交机器学习算法，例如Chernozhukov等人的双重机器学习工作。 该工具包旨在测量某些治疗变量t对结果变量y的因果关系，并控制一组特征x （197个子文件） 
data_doc.bib 274B setup.cfg 2KB example_full.csv 304KB card.csv 293KB example.csv 84KB sim.csv 65KB NEW7080.dta 37.54MB .gitignore 1KB Generalized Random Forests.ipynb 1.27MB Double Machine Learning Examples.ipynb 800KB TA_DGP_analysis.ipynb 651KB TA_DGP_analysis_Step_CATE.ipynb 649KB Interpretability with SHAP.ipynb 420KB NLSYM_Linear.ipynb 391KB ForestLearners Basic Example.ipynb 356KB high_dim_state_any_m_panel.ipynb 331KB high_dim_state_any_m_panel_hetero.ipynb 318KB Case Study - Recommendation AB Testing at An Online Travel Company - EconML + DoWhy.ipynb 309KB Automated Machine Learning For EconML.ipynb 303KB Case Study - Customer Segmentation at An Online Media Company - EconML + DoWhy.ipynb 294KB Causal Model Selection with the RScorer.ipynb 292KB Doubly Robust Learner and Interpretability.ipynb 280KB Case Study - Multi-investment Attribution at A Software Company - EconML + DoWhy.ipynb 264KB NLSYM_Semi_Synthetic_Linear.ipynb 262KB Causal Forest and Orthogonal Random Forest Examples.ipynb 250KB NLSYM_Semi_Synthetic_GBM.ipynb 249KB Case Study - Customer Segmentation at An Online Media Company.ipynb 245KB Case Study - Recommendation AB Testing at An Online Travel Company.ipynb 236KB Weighted Double Machine Learning Examples.ipynb 202KB NLSYM_GBM.ipynb 196KB postprocess_panel_hetero.ipynb 146KB Case Study - Multi-investment Attribution at A Software Company.ipynb 136KB Deep IV Examples.ipynb 129KB Metalearners Examples.ipynb 103KB Choosing First Stage Models.ipynb 88KB post_processing.ipynb 85KB postprocess_panel.ipynb 55KB LICENSE 3KB README.md 26KB README.md 3KB README.md 2KB README.md 1KB
dr_cate_tree.png 183KB dr_policy_tree.png 66KB continuous_ortho_forest_doc_example.png 30KB summary.png 28KB summary_frame.png 15KB population_summary.png 12KB _splitter.pxd 7KB _tree.pxd 6KB _criterion.pxd 5KB _criterion.pxd 5KB _utils.pxd 3KB _utils.pxd 1KB linear_model.py 76KB _drlearner.py 70KB dml.py 65KB _inference.py 65KB test_dml.py 62KB _ortho_forest.py 62KB test_drlearner.py 53KB _cate_estimator.py 53KB classes.py 52KB _base_grf.py 51KB test_statsmodels.py 49KB _dml.py 41KB _ortho_learner.py 41KB utilities.py 41KB test_grf_python.py 39KB _base_grftree.py 36KB _dr.py 35KB test_linear_model.py 35KB causal_forest.py 30KB test_deepiv.py 29KB _interpreters.py 29KB monte_carlo_statsmodels.py 27KB ortho_forest.py 26KB dr_iv.py 23KB test_orf.py 22KB _metalearners.py 21KB test_ortho_iv.py 20KB _rlearner.py 19KB test_ortho_learner.py 19KB model_selection.py 18KB _shap.py 18KB test_inference.py 18KB _deepiv.py 17KB test_refit.py 17KB test_grf_cython.py 16KB _tree_exporter.py 16KB test_bootstrap.py 15KB test_tree.py 15KB dml_iv.py 15KB dynamic_panel_dgp.py 15KB _automated_ml.py 15KB coverage_panel_hetero.py 15KB _tsls.py 15KB utilities.py 14KB comparison_plots.py 14KB monte_carlo.py 14KB共 197 条
- 1
- 2
评论0