Repository files navigation

The R packageDoubleML provides an implementation of the double /debiased machine learning framework ofChernozhukov etal. (2018). It is built on top ofmlr3 and themlr3ecosystem(Lang et al., 2019).

Note that the R package was developed together with a python twin basedonscikit-learn. The python package is alsoavailable onGitHub and.

Documentation of functions in R:https://docs.doubleml.org/r/stable/reference/index.html

User guide:https://docs.doubleml.org

DoubleML is currently maintained by@PhilippBach and@SvenKlaassen.

Double / debiased machine learning framework ofChernozhukov etal. (2018) for

• Partially linear regression models (PLR)
• Partially linear IV regression models (PLIV)
• Interactive regression models (IRM)
• Interactive IV regression models (IIVM)

The object-oriented implementation ofDoubleML that is based on theR6 package for R is very flexible. The modelclassesDoubleMLPLR,DoubleMLPLIV,DoubleMLIRM andDoubleIIVMimplement the estimation of the nuisance functions via machine learningmethods and the computation of the Neyman orthogonal score function. Allother functionalities are implemented in the abstract base classDoubleML. In particular functionalities to estimate double machinelearning models and to perform statistical inference via the methodsfit,bootstrap,confint,p_adjust andtune. Thisobject-oriented implementation allows a high flexibility for the modelspecification in terms of …

• … the machine learning methods for estimation of the nuisancefunctions,
• … the resampling schemes,
• … the double machine learning algorithm,
• … the Neyman orthogonal score functions,
• …

It further can be readily extended with regards to

• … new model classes that come with Neyman orthogonal score functionsbeing linear in the target parameter,
• … alternative score functions via callables,
• … alternative resampling schemes,
• …

Install the latest release from CRAN:

remotes::packages("DoubleML")

Install the development version from GitHub:

remotes::install_github("DoubleML/doubleml-for-r")

DoubleML requires

• R (>= 3.5.0)
• R6 (>= 2.4.1)
• data.table (>= 1.12.8)
• stats
• checkmate
• mlr3 (>= 0.5.0)
• mlr3tuning (>= 0.3.0)
• mlr3learners (>= 0.3.0)
• mvtnorm
• utils
• clusterGeneration
• readstata13

DoubleML is a community effort. Everyone is welcome to contribute. Toget started for your first contribution we recommend reading ourcontributingguidelinesand ourcode ofconduct.

If you use the DoubleML package a citation is highly appreciated:

Bach, P., Chernozhukov, V., Kurz, M. S., and Spindler, M. (2021),DoubleML - An Object-Oriented Implementation of Double Machine Learningin R, arXiv:2103.09603.

Bibtex-entry:

@misc{DoubleML2020,      title={{DoubleML} -- {A}n Object-Oriented Implementation of Double Machine Learning in {R}},       author={P. Bach and V. Chernozhukov and M. S. Kurz and M. Spindler and Sven Klaassen},      year={2024},      journal={Journal of Statistical Software},      volume={108},      number={3},      pages= {1-56},      doi={10.18637/jss.v108.i03},      note={arXiv:\href{https://arxiv.org/abs/2103.09603}{2103.09603} [stat.ML]}}

Funding by the Deutsche Forschungsgemeinschaft (DFG, German ResearchFoundation) is acknowledged – Project Number 431701914.

• Bach, P., Chernozhukov, V., Kurz, M. S., Spindler, M. and Klaassen, S.(2024), DoubleML - An Object-Oriented Implementation of Double MachineLearning in R, Journal of Statistical Software, 108(3): 1-56,doi:10.18637/jss.v108.i03,arXiv:2103.09603.

• Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C.,Newey, W. and Robins, J. (2018), Double/debiased machine learning fortreatment and structural parameters. The Econometrics Journal, 21:C1-C68,https://doi.org/10.1111/ectj.12097.

• Lang, M., Binder, M., Richter, J., Schratz, P., Pfisterer, F., Coors,S., Au, Q., Casalicchio, G., Kotthoff, L., Bischl, B. (2019), mlr3: Amodern object-oriented machine learing framework in R. Journal of OpenSource Software,https://doi.org/10.21105/joss.01903.