predict.psicaTree

object

<p>should the matrix or class probabilities be predicted (TRUE)
or the classes themselves (FALSE)</p>

prob

<p>further parameters to be passed to 'rpart' object.</p>

In the situation when multiple alternative treatments or
interventions available, different population groups may respond differently
to different treatments. This package implements a method that discovers
the population subgroups in which a certain treatment has a better effect
than the other alternative treatments. This is done by first estimating the
treatment effect for a given treatment and its uncertainty by computing random
forests, and the resulting model is summarized by a decision tree in which the
probabilities that the given treatment is best for a given subgroup is shown in
the corresponding terminal node of the tree.

predict.psicaTree: Makes predictions for PSICA tree.

Description

Usage

Arguments