With bounded potential outcomes, independent randomization plus a midpoint-centered, no-intercept regression is minimax for finite-population average treatment effects, and complete randomization is strictly suboptimal.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
econ.EM 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Optimal Experimental Design and Estimation when Potential Outcomes are Bounded
With bounded potential outcomes, independent randomization plus a midpoint-centered, no-intercept regression is minimax for finite-population average treatment effects, and complete randomization is strictly suboptimal.