arXiv 6 Nov 2025 · Statistics — Methodology
arXiv:2511.04658 · PDF · DOI · OpenAlex · Extracted main text
How should researchers select experimental sites when the deployment population differs from observed data? I formulate the problem of experimental site selection as an optimal transport problem, developing methods to minimize downstream estimation error by choosing sites that minimize the Wasserstein distance between population and sample covariate distributions. I develop new theoretical upper bounds on PATE and CATE estimation errors, and show that these different objectives lead to different site selection strategies. I extend this approach by using Wasserstein Distributionally Robust Optimization to develop a site selection procedure robust to adversarial perturbations of covariate information: a specific model of distribution shift. I also propose a novel data-driven procedure for selecting the uncertainty radius the Wasserstein DRO problem, which allows the user to benchmark robustness levels against observed variation in their data. Simulation evidence, and a reanalysis of a randomized microcredit experiment in Morocco (Crépon et al.), show that these methods outperform random and stratified sampling of sites when covariates have prognostic R-squared > .5, and alternative optimization methods i) for moderate-to-large size problem instances ii) when covariates are moderately informative about treatment effects, and iii) under induced distribution shift.
appendix boundary found by appendix_command · 65% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Naoki Egami and Diana Da In Lee (2024) Designing multi-site studies for external validity: Site selection via synthetic purposive sampling, 2024 | 0.928 | 5 | 3 | 80% |
| 2 | Clara Bicalho, Adam Bouyamourn, and Thad Dunning (2022) Conditional balance tests: Increasing sensitivity and specificity with prognostic covariates, 2022 self | 0.843 | 3 | 3 | 100% |
| 3 | Elizabeth Tipton (2013) Improving generalizations from experiments using propensity score subclassification: Assumptions, properties, and contexts | 0.737 | 3 | 2 | 100% |
| 4 | T Dunning, G Grossman, M Humphreys, S D Hyde, C McIntosh, and G Nell… (2019) Information, Accountability, and Cumulative Learning: Lessons from Metaketa I | 0.644 | 3 | 2 | 67% |
| 5 | Jacques E Rossouw, Garnet L Anderson, Ross L Prentice, Andrea Z LaCr… (2002) Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the women's health initiat… | 0.644 | 3 | 2 | 67% |
| 6 | Jose Blanchet and Karthyek Murthy (2018) Quantifying distributional model risk via optimal transport | 0.644 | 2 | 2 | 100% |
| 7 | Jennifer L. Hill (2010) Bayesian nonparametric modeling for causal inference | 0.644 | 2 | 2 | 100% |
| 8 | Frank H. Knight (1921) Risk, Uncertainty and Profit | 0.644 | 2 | 2 | 100% |
| 9 | Cass R. Sunstein (2023) Knightian uncertainty | 0.644 | 2 | 2 | 100% |
| 10 | Marco Cuturi (2013) Sinkhorn distances: Lightspeed computation of optimal transportation distances, 2013 | 0.644 | 2 | 2 | 100% |
Showing the top 10 of 130 scored citations.