I can’t run regression trees without geeglm. I have longitudinal data so rpart wouldn’t work. Is there a way to get regression trees with geeglm?
I'm not aware of any decision tree based on
geeglm. However, you can use the
glmtree() function from the
partykit package and specify the
cluster= argument. This essentially corresponds to using
geeglm() with an independence working model and a clustered covariance matrix estimate.
Alternatively, you can use
glmertree() from the package
glmertree which combines the
glmer() function from
lme4 for estimating the random effects and
partykit to estimate the tree structure. For more details see: Fokkema et al. (2018). "Detecting Treatment-Subgroup Interactions in Clustered Data with Generalized Linear Mixed-Effects Model Trees." Behavior Research Methods, 50(5), 2016-2034. doi:10.3758/s13428-017-0971-x