Open Access
Translator Disclaimer
March 2008 On regression adjustments in experiments with several treatments
David A. Freedman
Ann. Appl. Stat. 2(1): 176-196 (March 2008). DOI: 10.1214/07-AOAS143

Abstract

Regression adjustments are often made to experimental data. Since randomization does not justify the models, bias is likely; nor are the usual variance calculations to be trusted. Here, we evaluate regression adjustments using Neyman’s nonparametric model. Previous results are generalized, and more intuitive proofs are given. A bias term is isolated, and conditions are given for unbiased estimation in finite samples.

Citation

Download Citation

David A. Freedman. "On regression adjustments in experiments with several treatments." Ann. Appl. Stat. 2 (1) 176 - 196, March 2008. https://doi.org/10.1214/07-AOAS143

Information

Published: March 2008
First available in Project Euclid: 24 March 2008

zbMATH: 1144.62027
MathSciNet: MR2415599
Digital Object Identifier: 10.1214/07-AOAS143

Keywords: balanced designs , bias , estimation , experiments , intention-to-treat , models , multiple regression , Randomization

Rights: Copyright © 2008 Institute of Mathematical Statistics

JOURNAL ARTICLE
21 PAGES


SHARE
Vol.2 • No. 1 • March 2008
Back to Top