Artificial dataset for illustrating (G)LMM trees for detecting treatment-subgroup interactions.

1 | ```
data("TreatmentSubgroups")
``` |

A data frame containing 1000 observations on 8 variables.

- u1,u2,u3,u4,u5
numeric. Continuous partitioning variables.

- treatment
factor. Binary treatment variable.

- cluster
factor. Indicator for cluster with 10 levels.

- ynum
numeric. Continuous treatment outcome.

- ybin
factor. Binary treatment outcome.

The data contains four treatment effect subgroups with respect to the
continuous partitioning variables (`u1`

–`u5`

). The `cluster`

variable has an additional random intercept that should be accounted for.
The outcome is assessed either by a continuous variable (`ynum`

) or
a binary variable (`ybin`

).

There are two large subgroups where Treatment 1 is better than Treatment 2 and
vice versa, respectively. Additionally, there are two smaller subgroups where
both treatments lead to comparable outcomes. For the corresponding (G)LMM trees
see `lmertree`

.

`lmertree`

1 2 | ```
data("TreatmentSubgroups", package = "glmertree")
summary(TreatmentSubgroups)
``` |

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.