Adaptive Experimental Design Using Shrinkage Estimators
Evan T. R. Rosenman, Kristen B. Hunter · 2026
Read Adaptive Experimental Design Using Shrinkage Estimators by Evan T. R. Rosenman, Kristen B. Hunter (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-07404.txt.
Description
Adaptive trials assign treatments as results arrive, aiming to estimate effects efficiently or find the best treatment. Evan T. R. Rosenman and one co-author improve that adaptivity by borrowing strength across arms with Stein-like shrinkage estimators.
Classical Neyman-style adaptive allocation minimizes error for a Horvitz-Thompson estimator but ignores information shared across treatment arms. Shrinkage estimators use that shared information, cutting expected squared error against estimating each effect alone. Their risk can be computed efficiently as expectations of ratios of Gaussian quadratic forms. Trial builders get adaptive designs that learn from every arm at once.
Catalog record
| CREATOR | Evan T. R. Rosenman, Kristen B. Hunter |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 99 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-07404