What is it about?
Compares Athony Atkinson's design to minimimation in the case where there is balancing using a number of binary covariates
Featured Image
Why is it important?
Provides many theoretical results as well as the results of simulation. Uses a unique device for generating covariate patterns with different degrees of covariate dependence. Producers a number of key insights regarding losses of randomisation. Shows that in practice these losses are small.
Perspectives
This paper was inspired by a claim that binary covariates provide an exception to Atkinson's algorithm being seuperior to minimisation. On investigation this turned out not to be true.
Professor Stephen J Senn
Consultant Statistician
Read the Original
This page is a summary of: Comparisons of minimization and Atkinson's algorithm, Statistics in Medicine, March 2010, Wiley,
DOI: 10.1002/sim.3763.
You can read the full text:
Resources
Contributors
The following have contributed to this page