|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|||||||||||
|
|
|
|
||||||||||||||||
|
|
||||||||||||||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|||||||
|
|
|
|||||||||||||||||
|
|
ABSTRACT Margaret Wu, National Institutes of Health Different approaches to account for informative
missingness in repeated measures are reviewed. We
then discuss how to apply the conditional model
approach (Wu and Bailey 1989 and Follmann and Wu
1995) to the setting where the probability of missing
a visit depends on the random effects in a time
dependent fashion. This includes the case where the
probability of missing a visit depens on the true value
of the primary response. Summary statistics for
missingness that are weighted sums of the missing
indicators are derived for these situations. These
summary statistics are then incorporated as fixed
effect covariates in the random effects model for the
primary response. Applications of these summary
statistics are then illustrated by analyzing data from
real life examples.
Return to Longitudinal/Survival Working Group List | Return to Home Page |
