Metraipolis › archive › dc_d30b2d0ed226
Studentized inference in randomized experiments when a few units dominate the outcome variation — arXiv:2611.03561
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.03561
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:44:25.825Z
- SHA-256 of the archived text
69358c108976c1c2e1f327f2f17b8a514a7ebe073ea1f533606b2f48d42ff62e- Stable URL of this page
https://metraipolis.com/documento/dc_d30b2d0ed226
How to cite
Studentized inference in randomized experiments when a few units dominate the outcome variation — arXiv:2611.03561 — arXiv (en), https://arxiv.org/abs/2611.03561 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d30b2d0ed226 (Metraipolis archive dc_d30b2d0ed226, sha256 69358c108976c1c2e1f327f2f17b8a514a7ebe073ea1f533606b2f48d42ff62e)
BibTeX
@misc{dcd30b2d0ed226,
title = {Studentized inference in randomized experiments when a few units dominate the outcome variation — arXiv:2611.03561},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.03561},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d30b2d0ed226; sha256 69358c108976c1c2e1f327f2f17b8a514a7ebe073ea1f533606b2f48d42ff62e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03561. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In completely randomized experiments with heavy-tailed outcomes, a few units can carry a share of the residual variation that does not vanish as the number of units grows. We study the studentized difference in means and Lin's regression-adjusted estimator, with covariates whose leverage vanishes, when the finite population is fixed, and the treatment assignment is the only source of randomness. The studentized statistic then has a nonnormal limit in general. Populations with the same marginal distributions of potential outcomes can have different limits, as the limit depends on the unobserved pairing mechanism of the two potential outcomes of each dominant unit. When the outcome errors are drawn from a heavy-tailed law, the limit is random and, outside a degenerate case, varies with the population. Under balanced allocation, the conventional Wald interval has an explicit asymptotic coverage floor. At the nominal level $0.95$, the floor is about $0.42$ for every population and about $0.84$ when the two potential residuals of each unit are equal. Both floors are close to the best possible, and without balanced allocation, the coverage can fall to the smaller of the proportions of treated and control units. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv stat.ME, nuovi lavori del giorno). Categorie arXiv: Statistics Theory (math.ST) ; Methodology (stat.ME). Pagina del lavoro: https://arxiv.org/abs/2611.03561 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T18:13:53.158Z dalla pagina https://arxiv.org/list/stat.ME/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).