Double/Debiased Machine Learning for Treatment and Causal Parameters

By Victor Chernozhukov, D. Chetverikov, Mert Demirer, Esther Duflo, Christian Hansen, Whitney K. Newey, James M. Robins. 2016

158 citations recorded by the source corpus.

Most modern supervised statistical/machine learning (ML) methods are explicitly designed to solve prediction problems very well. Achieving this goal does not imply that these methods automatically deliver good estimators of causal parameters. Examples of such parameters include individual regression coefficients, average treatment effects, average lifts, and demand or supply elasticities. In fact, estimates of such causal parameters obtained via naively plugging ML estimators into estimating equations for such parameters can behave very poorly due to the regularization bias. Fortunately, this regularization bias can be removed by solving auxiliary prediction problems via ML tools. Specifically, we can form an orthogonal score for the target low-dimensional parameter by combining auxiliary and main ML predictions. The score is then used to build a de-biased estimator of the target parameter which typically will converge at the fastest possible 1/root(n) rate and be approximately unbiased and normal, and from which valid confidence intervals for these parameters of interest may be constructed. The resulting method thus could be called a "double ML" method because it relies on estimating primary and auxiliary predictive models. In order to avoid overfitting, our construction also makes use of the K-fold sample splitting, which we call cross-fitting. This allows us to use a very broad set of ML predictive methods in solving the auxiliary and main prediction problems, such as random forest, lasso, ridge, deep neural nets, boosted trees, as well as various hybrids and aggregators of these methods.

Subfields: Statistics and Probability, Causal Inference & Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

In Causal Inference & Discovery

Citations in the corpus

The hardness of conditional independence testing and the generalised covariance measure (2018, methodology): “Estimation of expected conditional covariance Though typically not thought of as conditional independence tests, there are several approaches to estimating the expected conditional covariance functional Ecov(X,Y |Z) in the semiparametric statistics literature [46, 15, 40].”
Double/De-Biased Machine Learning of Global and Local Parameters Using Regularized Riesz Representers (2018, methodology): “…chosen proportionally to √ log(p ∨ n)/√n, that is ℓ̃ ≍ √ log(p ∨ n) using self-normalized moderate deviation bounds (Jing et al. (2003); Belloni et al. (2014)) or high-dimensional central limit theorems (Chernozhukov et al. (2017)), under mild moment conditions, without requiring sub-Gaussianity.”
Estimation and Inference about Heterogeneous Treatment Effects in High-Dimensional Dynamic Panels. (2017, background): “1 from Chernozhukov et al. (2017a) returns unconditional bound: ‖āk‖ = sup α∈Sd−1 ‖ākα‖ .”
Overlap in observational studies with high-dimensional covariates (2017, background): “…unconfoundedness and overlap in the presence of many covariates is particularly relevant in light of recent methodological developments that incorporate machine learning methods in semiparametric causal effect estimation (van der Laan & Gruber, 2010; Chernozhukov et al., 2016; Athey et al., 2016).”
Confounding-Robust Policy Improvement (2018, background): “Wager and Athey (2017a) show that since this estimate is semiparametrically efficient when using cross-fold fitting, as shown by Chernozhukov et al. (2016), this leads to better regret bounds.”