High-dimensional regression adjustments in randomized experiments
Authors:
Preprint
Abstract
We study the problem of treatment effect estimation in randomized experiments with high-dimensional regression adjustments. We present an inferential approach that takes any risk-consistent regression adjustment and yields efficient estimates of the average treatment effect. This considerably extends the range of settings where high-dimensional regression adjustments are guaranteed to provide valid inference about the population average treatment effect. We also 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, or many other methods; we also extend our analysis to allow for adaptive specification search via cross-validation, and flexible non-parametric regression adjustments with, e.g., random forests or neural networks.Keywords
All Related Versions
This publication has 32 references indexed in Scilit:
- Decomposing Treatment Effect VariationJournal of the American Statistical Association, 2018
- Estimation and Inference of Heterogeneous Treatment Effects using Random ForestsJournal of the American Statistical Association, 2018
- Recursive partitioning for heterogeneous causal effectsProceedings of the National Academy of Sciences of the United States of America, 2016
- Lasso adjustments of treatment effect estimates in randomized experimentsProceedings of the National Academy of Sciences, 2016
- Robust inference on average treatment effects with possibly more covariates than observationsJournal of Econometrics, 2015
- Spectral Analysis of Large Dimensional Random Matrices2010
- On regression adjustments to experimental dataAdvances in Applied Mathematics, 2008
- The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trialsStatistics in Medicine, 2006
- Estimating causal effects of treatments in randomized and nonrandomized studies.Journal of Educational Psychology, 1974
- DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICESMathematics of the USSR-Sbornik, 1967