1Department of Statistics, Columbia University, New York, NY, 10027, USA. gelman@stat.columbia.edu
For estimating linear treatment effects, optimal designs typically use extreme points. However, for nonlinear dose-response relationships, avoiding intermediate points is often best unless sample sizes are very large.
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Area of Science:
Background:
Purpose of the Study:
Main Methods:
Main Results:
Conclusions: