Pathwise Estimation of Covariate Balancing Propensity Scores

🚧 Work in progress 🚧

A package for pathwise estimation of regularized logistic propensity score models using covariate balancing loss functions rather than maximum likelihood. Regularization paths are fit via the adelie elastic-net solver with an interface inspired by glmnet, and objectives that directly target covariate balance for the ATE and ATT.

Some helpful links for getting started:

• An introduction to balnet
• Package docs

Installation

The development version can be installed via

devtools::install_github("erikcs/balnet", subdir = "r-package/balnet")

Installing from source requires a C++17 compiler or later. To build with multithreading enabled, OpenMP needs to be available (on Mac, a simple option is to set the default C++ compiler to gcc installed via brew).