High-dimensional regression adjustments in randomized experiments

By Stefan Wager, Wenfei Du, Jonathan E. Taylor, Robert Tibshirani. 2016

127 citations recorded by the source corpus.

Significance As datasets get larger and more complex, there is a growing interest in using machine-learning methods to enhance scientific analysis. In many settings, considerable work is required to make standard machine-learning methods useful for specific scientific applications. We find, however, that in the case of treatment effect estimation with randomized experiments, regression adjustments via machine-learning methods designed to minimize test set error directly induce efficient estimates of the average treatment effect. Thus, machine-learning methods can be used out of the box for this task, without any special-case adjustments. We study the problem of treatment effect estimation in randomized experiments with high-dimensional covariate information and show that essentially any risk-consistent regression adjustment can be used to obtain efficient estimates of the average treatment effect. Our results considerably extend the range of settings where high-dimensional regression adjustments are guaranteed to provide valid inference about the population average treatment effect. We then propose cross-estimation, a simple method for obtaining finite-sample–unbiased treatment effect estimates that leverages high-dimensional regression adjustments. Our method can be used when the regression model is estimated using the lasso, the elastic net, subset selection, etc. Finally, we extend our analysis to allow for adaptive specification search via cross-validation and flexible nonparametric regression adjustments with machine-learning methods such as random forests or neural networks.

Subfields: Statistics and Probability

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

Citations in the corpus

Quasi-oracle estimation of heterogeneous treatment effects (2017, methodology): “We also note that using held-out prediction for nuisance components, also known as cross-fitting, is an increasingly popular approach for making machine learning methods usable in classical semiparametrics (Athey and Wager, 2017; Chernozhukov et al., 2018; Schick, 1986; van der Laan and Rose, 2011; ”
Rerandomization and regression adjustment (2019, methodology): “It is interesting to extend the discussion to covariate adjustment in more complicated settings, such as high dimensional covariates (Bloniarz et al. 2016; Wager et al. 2016; Lei and Ding 2018), logistic regression for binary outcomes (Zhang et al.”
Covariate powered cross‐weighted multiple testing (2017, background): “ion of (convex) structural constraints on the final decision boundary. 20 rows. Furthermore, cross-weighting is similar to cross-fitting (Chernozhukov et al., 2017; Nie and Wager, 2018; Schick, 1986; Wager et al., 2016), a method popular in the field of causal inference for dealing with high-dimensi”
Covariate-adjusted Fisher randomization tests for the average treatment effect (2020, background): “2019; Guo and Basse 2020), and even estimators based on machine learning algorithms (Wager et al. 2016; Wu and Gagnon-Bartsch 2018; Chen et al. 2020).”
The Generalized Oaxaca-Blinder Estimator (2020, background): “If one is willing to assume that (xi, y0i, y1i, Zi) are iid samples from an infinite superpopulation, then clever use of sample splitting can circumvent these assumptions (Wager et al. 2016; Chernozhukov et al. 2018).”
Policy Learning With Observational Data (2017, background): “…uses the true per-county treatment fractions ˆ e ∗ i and estimates nuisance components using cross-ﬁtting, the point estimates will be √ n -consistent and the associated conﬁdence intervals asymptotically valid essentially without assumptions (Rothe, 2018; Wager, Du, Taylor, and Tibshirani, 2016).”