Causal ML
Methods that combine flexible machine learning with causal identification, enabling valid effect estimation in high-dimensional settings and principled estimation of treatment effect heterogeneity.
Classical causal methods require the researcher to specify control variables parametrically. In high-dimensional settings this introduces omitted variable bias and model misspecification.
ML models can approximate complex nuisance functions, the conditional mean of Y or T given X, without specifying a parametric form. But naive ML estimates are regularization-biased for causal parameters.
Causal ML resolves this tension: use flexible ML for the parts you don't care about (nuisance), and apply orthogonalization and cross-fitting to recover valid, root-n consistent causal estimates.
library(DoubleML)
# The doubly robust (AIPW) score, with cross-fitting
obj <- DoubleMLIRM$new(
data = dml_data,
ml_g = lrn("regr.ranger"), # outcome model
ml_m = lrn("classif.ranger", predict_type="prob"), # propensity model
n_folds = 5,
score = "ATE"
)
obj$fit()
obj$summary()Causal forest
Extends random forests to estimate heterogeneous treatment effects (CATE) at the individual level. Uses honest splitting and local centering to debias estimates across the covariate space. Also maps the wider meta-learner landscape: S, T, X, R, and DR learners.
AIPW
Augmented inverse probability weighting combines outcome and propensity models into a doubly robust score, consistent if either nuisance model is correctly specified. Efficient semiparametrically.