WhyNot Estimators

A collection of causal inference estimators implemented in Python and R to pair with the Python causal inference library whynot. For more information, check out the documentation.

Installation

You can perform a minimal installation of whynot_estimators with

git clone https://github.com/zykls/whynot_estimators.git
cd whynot_estimators
pip install -r requirements.txt

You can also install via pip

pip install whynot_estimators

This installs the basic framework. Additional estimators, along with their dependencies are installed separately. To see a list of all available estimators, use

python -m whynot_estimators show_all

To install a particular estimator, e.g. the causal_forest, run

python -m whynot_estimators install causal_forest

Note this estimator requires a working R installation. The show_all command also shows which estimators require R.

To install all of the estimators, use

python -m whynot_estimators install all

Alternatively, you can install the dependencies for a specific estimator by hand by looking here.

Installing R

Some estimators in whynot_estimators require a functioning R installation. One way to satisfy this requirement is using conda.

# Install Anaconda (Linux)
wget https://repo.anaconda.com/archive/Anaconda3-2018.12-Linux-x86_64.sh
bash Anaconda3-2018.12-Linux-x86_64.sh

# Install Anaconda (MacOSx)
wget https://repo.anaconda.com/archive/Anaconda3-2019.03-MacOSX-x86_64.sh
bash Anaconda3-2019.03-MacOSX-x86_64.sh

# Create R environment
conda create --name whynot r-essentials r-base