An Algorithm for the Univariate Analysis of Variance in Experiments with Repeated Measures.
Final rept. Aug 80-Aug 82,
SCHOOL OF AEROSPACE MEDICINE BROOKS AFB TX
Pagination or Media Count:
A computing procedure is described for the univariate analysis of a repeated measurements experiment where the experimental units frequently subjects are arranged in a two-way classification with cell frequencies that can be disproportionate. The analysis is adjusted for missing values, provided their number and configuration do not violate certain limitations. Unlike some strategies for handling missing values in repeated measurements experiments, the method does not require the inclusion in the model of an explicit subject factor, meaning that the order of the matrix to be inverted does not depend on the number of subjects in the experiment. The algorithm has been incorporated into a SAS procedure, REP2W1F, which computes the full analysis with a single call and produces useful summary statistics including least-squares means particular to the design. The approach could be generalized to experiments where the number of treatment factors is other than two and where the repeated measures that a factorial arrangement of their own.
- Statistics and Probability
- Computer Programming and Software
- Test Facilities, Equipment and Methods