AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Synthetic Design: An Optimization Approach to Experimental Design with Synthetic Controls

Nick Doudchenko, Khashayar Khosravi, Jean Pouget-Abadie, Sebastien Lahaie, Miles Lubin, Vahab Mirrokni, Jann Spiess, Guido Imbens

NeurIPS 2021 · Learning theory and algorithms › Combinatorial and search optimization · approach: Predict-then-optimize learning · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
optimize treated units with per-unit design instead of random synthetic controlsynthetic control · BLS simulation, heterogeneous, K=3unit-level RMSE (x10^3): 15.9 -> 13.9Improved
optimize treated units with per-unit design instead of random synthetic controlsynthetic control · BLS simulation, heterogeneous, K=7unit-level RMSE (x10^3): 19.0 -> 16.0Improved
per-unit design instead of random synthetic controlsynthetic control · BLS simulation, homogeneous, K=3ATET RMSE (x10^3): 9.7 -> 8.5Improved
per-unit design instead of random synthetic controlsynthetic control · BLS simulation, homogeneous, K=7ATET RMSE (x10^3): 10.3 -> 8.3Improved
synthetic control with random treatment instead of random diff-in-meansdiff-in-means · BLS simulation, homogeneous, K=3ATET RMSE (x10^3): 12.1 -> 9.7Improved
two-way global design instead of diff-in-meansdiff-in-means · BLS simulation, homogeneous, K=3ATET RMSE (x10^3): 12.1 -> 8.4Improved
two-way global design instead of per-unit designper-unit · BLS simulation, homogeneous, K=3unit-level RMSE (x10^3): 13.9 -> 8.4Improved
two-way global design instead of per-unit designper-unit · BLS simulation, heterogeneous, K=3unit-level RMSE (x10^3): 13.9 -> 27.6Hurt
two-way global design instead of per-unit designper-unit · BLS simulation, heterogeneous, K=7unit-level RMSE (x10^3): 16.0 -> 32.5Hurt
one-way instead of two-way global designtwo-way global · BLS simulation, homogeneous, K=3ATET RMSE (x10^3): 8.4 -> 8.5No clear effect
two-way global design in heterogeneous setting vs per-unitper-unit · BLS simulation, heterogeneous, K=7ATET RMSE (x10^3): 8.3 -> 8.9Hurt
increase treated units from K=3 to K=7 for synthetic controlsynthetic control · BLS simulation, homogeneousunit-level RMSE (x10^3): 15.9 -> 19.0Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy