dmlalg {dmlalg}

dmlalg: double machine learning algorithms

Description

The dmlalg package contains implementations of double machine learning (DML) algorithms in R.

Partially linear models with confounding variables

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.

regsdml

Estimates the linear parameter in a partially linear model with confounding variables with regularized and standard DML methods.

summary.regsdml

A summary method for objects fitted with regsdml.

confint.regsdml

A confint method for objects fitted with regsdml.

coef.regsdml

A coef method for objects fitted with regsdml.

vcov.regsdml

A vcov method for objects fitted with regsdml.

print.regsdml

A print method for objects fitted with regsdml.

Partially linear mixed-effects models with repeated measurements

Our goal is to estimate and perform inference for the linear coefficient in a partially linear mixed-effects model with DML. Machine learning algorithms allows us to incorporate more complex interaction structures and high-dimensional variables.

mmdml

Estimates the linear parameter in a PLMM with repeated measurements using double machine learning.

confint.mmdml

A confint method for objects fitted with mmdml.

fixef.mmdml

A fixef method for objects fitted with mmdml.

print.mmdml

A print method for objects fitted with mmdml.

ranef.mmdml

A ranef method for objects fitted with mmdml.

residuals.mmdml

A residuals method for objects fitted with mmdml.

sigma.mmdml

A sigma method for objects fitted with mmdml.

summary.mmdml

A summary method for objects fitted with mmdml.

vcov.mmdml

A vcov method for objects fitted with mmdml.

VarCorr.mmdml

A VarCorr method for objects fitted with mmdml.

References

C. Emmenegger and P. Bühlmann. Regularizing Double Machine Learning in Partially Linear Endogenous Models, 2021. Preprint arXiv:2101.12525.

C. Emmenegger and P. Bühlmann. Double Machine Learning for Partially Linear Mixed-Effects Models with Repeated Measurements. Preprint arXiv:2108.13657.