What is it about?
Many tests for changes in error variance structure do not extend easily beyond the one way layout. This test for heteroscedasticity extends to more complicated experimental designs and does not require the design to be balanced.
Featured Image
Why is it important?
The test for heteroscedasticity is exact making it important to use when there are not many replications of the experiment.
Perspectives
Read the Original
This page is a summary of: Uncorrelated residuals and an exact test for two variance components in experimental design, Communication in Statistics- Theory and Methods, January 1992, Taylor & Francis,
DOI: 10.1080/03610929208830927.
You can read the full text:
Contributors
The following have contributed to this page