PubMed · 8068847
M-estimation in cross-over trials.
Abstract
A robust procedure, combined M-estimation, is proposed for analyzing cross-over data with possible within- and between-subject outliers. The mean squared error properties of these combined M-estimates for direct treatment effect contrasts and carryover treatment effect contrasts are examined through simulation studies. They are found to be superior to those of the generalized least squares estimates when there are possible within- and between-subject outliers.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
E M Chi. 1994. M-estimation in cross-over trials.. https://pubmed.ncbi.nlm.nih.gov/8068847/
Cite the original work for its findings. Save a collection to share your selection of sources.