Metraipolis › archive › dc_a74c4e6dda74

Empirical Bayes for compound adaptive experiments — arXiv:2609.17158

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.17158
License (as archived)
CC0-1.0
Archived
2026-10-15T22:38:38.872Z
SHA-256 of the archived text
d9aca501690e4d05c745dffd9f934b6841cc1cef40d80280bf10eb80cc1009be
Stable URL of this page
https://metraipolis.com/documento/dc_a74c4e6dda74

How to cite

Empirical Bayes for compound adaptive experiments — arXiv:2609.17158 — arXiv (en), https://arxiv.org/abs/2609.17158 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a74c4e6dda74 (Metraipolis archive dc_a74c4e6dda74, sha256 d9aca501690e4d05c745dffd9f934b6841cc1cef40d80280bf10eb80cc1009be)
BibTeX
@misc{dca74c4e6dda74,
  title = {Empirical Bayes for compound adaptive experiments — arXiv:2609.17158},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.17158},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a74c4e6dda74; sha256 d9aca501690e4d05c745dffd9f934b6841cc1cef40d80280bf10eb80cc1009be},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17158. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We investigate Empirical Bayes (EB) methods in the context of compound adaptive experiments, where the arm distribution in each experiment follows a normal distribution with an unknown mean that we seek to estimate. There are two main EB strategies: $g$-modeling, which estimates the prior by maximizing the marginal likelihood, and $f$-modeling, which derives posterior means directly from the empirical distribution of the observations. We show that $g$-modeling continues to be a valid EB procedure even when it incorrectly assumes that data are collected exogenously; its validity does not depend on the particular sampling algorithm or on whether sample sizes are endogenous. In practice, one can apply standard $g$-modeling techniques by acting as though the data were exogenously sampled. We extend regret guarantees from exogenous sampling to adaptively generated data. By contrast, naively applying the Tweedie formula based on the marginal density of the observed data, as in standard $f$-modeling, can produce biased rules under adaptive sampling. We corroborate the robustness of $g$-modeling through simulations with widely used adaptive algorithms and demonstrate its applicability using a real-world dataset consisting of multiple sequential experiments.

Tema degli esploratori: economia dei servizi IA (retro-catalogo arXiv: annunci del 2026-09-16 in econ.EM).
Categorie arXiv: Econometrics (econ.EM) ; Methodology (stat.ME).
Pagina del lavoro: https://arxiv.org/abs/2609.17158 — 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 2026-10-15T22:15:10.511Z dalla pagina https://arxiv.org/catchup/econ.EM/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).