Learning L2 Continuous Regression Functionals via Regularized Riesz Representers

Wednesday, October 16, 2019 - 4:00pm to 4:30pm

Event Calendar Category

LIDS & Stats Tea

Speaker Name

Rahul Singh



Building and Room Number

LIDS Lounge


Many objects of interest can be expressed as an L2 continuous functional of regression, including average treatment effects, economic average consumer surplus, expected conditional covariances, and discrete choice parameters that depend on expectations. Debiased machine learning (DML) of these objects requires a learning a Riesz representer (RR). We provide here Lasso and Dantzig learners of the RR and corresponding learners of affine and other nonlinear functionals. We give an asymptotic variance estimator for DML. We allow for a wide variety of regression learners that can converge at relatively slow rates. We give conditions for root-n consistency and asymptotic normality of the functional learner. We give results for non-affine functionals in addition to affine functionals.


Rahul is a 3rd year PhD in economics and statistics at MIT advised by Victor Chernozhukov, Anna Mikusheva, and Whitney Newey. Previously, he studied computational statistics and machine learning at the UCL Gatsby Unit. He works on projects in causal inference and statistical learning theory.