Double/debiased machine learning for treatment effects, Victor Chernozhukov, Denis Chetverikov, Mert Demirer, Esther Duflo, Christian Hansen, Whitney Newey, James Robins, 2018American Economic Review, Vol. 108DOI: 10.1257/aer.20170829 - Introduces Double Machine Learning (DML), a method for estimating causal effects that combines machine learning with Neyman orthogonality to achieve robust, asymptotically normal inference in high-dimensional settings.
High-dimensional penalized least squares in economics, Alexandre Belloni, Victor Chernozhukov, Christian Hansen, 2014Econometrica, Vol. 82 (Wiley-Blackwell)DOI: 10.3982/ECTA10729 - Discusses the use of Lasso and other penalized regression methods for estimating high-dimensional nuisance functions in econometric models, relevant for robust causal inference.
Causality, Judea Pearl, 2009 (Cambridge University Press)DOI: 10.1017/CBO9780511803161 - A foundational text on causal inference, introducing graphical models, the do-calculus, and criteria for identifying causal effects, including the backdoor criterion for adjustment sets. (2nd edition)