A repeated-measurements study is one in which subjects (animals, people, plots, etc) are observed on several occasions. Each subject usually receives some randomly allocated treatment, either at the outset or repeatedly through the investigation, and is then observed at successive occasions to see how the treatment effects develop. One way to analyse data sets like this is to use Genstat’s REML facilities to model the correlation structure over time.
REML |
fits a variance-component model by residual (or restricted) maximum likelihood |
|---|---|
VCOMPONENTS |
defines the model for REML |
VSTRUCTURE |
defines a variance structure for random effects in a REML model |
Alternatively, Genstat has procedures for customized plotting of the observations (or profiles) against time, repeated measures analysis of variance, analyses based on ante-dependence structure or generalized estimating equations, and regression or nonlinear modelling of data where the residuals follow an AR1 or power-distance correlation model.
ANTORDER |
assesses order of ante-dependence for repeated measures data |
|---|---|
ANTTEST |
calculates overall tests based on a specified order of ante-dependence |
AREPMEASURES |
produces an analysis of variance for repeated measurements |
CUMDISTRIBUTION |
fits frequency distributions to accumulated counts |
DREPMEASURES |
plots profiles and differences of profiles for repeated measurements |
GEE |
fits models to longitudinal data by generalized estimating equations |
NLAR1 |
fits curves with an AR1 or a power-distance correlation model |
RAR1 |
fits regressions with an AR1 or a power-distance correlation model |
VORTHPOLYNOMIAL |
calculates orthogonal polynomial time-contrasts for repeated measurements |