dmlalg: Double Machine Learning Algorithms

Implementation of double machine learning (DML) algorithms in R, based on Emmenegger and Buehlmann (2021) "Regularizing Double Machine Learning in Partially Linear Endogenous Models" <arXiv:2101.12525>. Our goal is to perform inference for the linear parameter in partially linear models with confounding variables. The standard DML estimator of the linear parameter has a two-stage least squares interpretation, which can lead to a large variance and overwide confidence intervals. We apply regularization to reduce the variance of the estimator, which produces narrower confidence intervals that are approximately valid. Nuisance terms can be flexibly estimated with machine learning algorithms.

Version: 0.0.2
Depends: R (≥ 4.0.0)
Imports: glmnet, matrixcalc, stats, splines, randomForest
Suggests: testthat (≥ 3.0.0)
Published: 2021-06-21
Author: Corinne Emmenegger ORCID iD [aut, cre], Peter Buehlmann ORCID iD [ths]
Maintainer: Corinne Emmenegger <emmenegger at stat.math.ethz.ch>
License: GPL (≥ 3)
URL: https://gitlab.math.ethz.ch/ecorinne/dmlalg.git
NeedsCompilation: no
Citation: dmlalg citation info
Materials: README NEWS
CRAN checks: dmlalg results

Downloads:

Reference manual: dmlalg.pdf
Package source: dmlalg_0.0.2.tar.gz
Windows binaries: r-devel: dmlalg_0.0.2.zip, r-release: dmlalg_0.0.2.zip, r-oldrel: dmlalg_0.0.2.zip
macOS binaries: r-release (arm64): dmlalg_0.0.2.tgz, r-release (x86_64): dmlalg_0.0.2.tgz, r-oldrel: dmlalg_0.0.2.tgz
Old sources: dmlalg archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=dmlalg to link to this page.