On regression adjustments to experimental data

Regression adjustments are often made to experimental data. Since randomization does not justify the models, almost anything can happen. Here, we evaluate results using Neyman's non-parametric model, where each subject has two potential responses, one if treated and the other if untreated. Only...

Full description

Saved in:
Bibliographic Details
Published inAdvances in applied mathematics Vol. 40; no. 2; pp. 180 - 193
Main Author Freedman, David A.
Format Journal Article
LanguageEnglish
Published San Diego, CA Elsevier Inc 01.02.2008
Elsevier
Subjects
Balance
Exact sciences and technology
General mathematics
General, history and biography
Intention-to-treat
Mathematics
Multiple regression
Numerical analysis
Numerical analysis. Scientific computation
Randomization
Sciences and techniques of general use
Randomization
62A01
Models
Balance
62J99
Multiple regression
Intention-to-treat
Asymptotic expansion
Error estimation
Applied mathematics
62J99. Models
Online AccessGet full text

Cover

More Information
Summary:Regression adjustments are often made to experimental data. Since randomization does not justify the models, almost anything can happen. Here, we evaluate results using Neyman's non-parametric model, where each subject has two potential responses, one if treated and the other if untreated. Only one of the two responses is observed. Regression estimates are generally biased, but the bias is small with large samples. Adjustment may improve precision, or make precision worse; standard errors computed according to usual procedures may overstate the precision, or understate, by quite large factors. Asymptotic expansions make these ideas more precise.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0196-8858
1090-2074
DOI:10.1016/j.aam.2006.12.003