Minimax Robust Designs for M-Estimated Models

Rui Hu, Douglas P. Wiens · 2026

Read Minimax Robust Designs for M-Estimated Models by Rui Hu, Douglas P. Wiens (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-2604-21998.txt.

Description

Hu and Wiens extend minimax experimental design theory from least squares to M-estimation, gaining robustness against outlying or erroneous responses. With a minor tweak to a tuning constant, least-squares-optimal designs stay optimal asymptotically for M-estimation. The authors argue even this change should be ignored, with the tuning constant chosen in a sensible way independent of the M-estimate used. Designs and estimates from independent, identically distributed errors also hold up, in a minimax sense, against broad correlation classes.

Catalog record

CREATORRui Hu, Douglas P. Wiens
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE59676
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.21998

Browse the full library · About the Tower of Babel