Description
Results from a single step of the BOWL algorithm.Objects from the Class
Objects can be created by calls of the form new("BOWL", ...).
These objects are for convenience in package development
and should not be created by users.Slots
shift: - An object of class
"numeric."
Rewards (or sum of rewards) must be
positive. This slot stores the
amount rewards (or sum of rewards) was
shifted to make all positive. step: - An object of class
"integer."
The step of the BOWL algorithm to which
this object pertains. ind: - An object of class
"logical."
The indicator of adherence to the optimal
treatment regime from this step through
final decision point. sumR: - An object of class
"numeric."
The unshifted total reward from this step through
final decision point. prodPr: - An object of class
"numeric."
The product of propensities scores
for treatment received from this step
through final
decision point. txInfo: - An object of class
"TxInfoBasic."
Treatment information. optim: - An object of class
"BOWLObj."
The optimization result(s).
Methods
- cvInfo
signature(object = "BOWL"):
Retrieve cross-validation results. - DTRstep
signature(object = "BOWL"):
Retrieve description of method used to create object. - optimObj
signature(object = "BOWL"):
Retrieve optimization results. - optTx
signature(x = "BOWL", newdata = "data.frame"):
Estimate optimal treatment for newdata at the decision
point for which the provided BOWL analysis pertains. - optTx
signature(x = "BOWL", newdata = "missing"):
Retrieve optimal treatment for training data at the decision
point for which the provided BOWL analysis pertains. - print
signature(x = "BOWL"):
Print key results of method. - regimeCoef
signature(object = "BOWL"):
Retrieve parameter estimates for regime. - show
signature(object = "BOWL"):
Show key results of method. - summary
signature(object = "BOWL"):
Retrieve key summary information of method.