Description
Results for a single step of the BOWL algorithm for
a subset of data with binary treatment options and no
subsets for propensity modeling.Objects from the Class
Objects can be created by calls of the form new("BOWLBasic", ...).
These objects are for convenience in package development
and should not be created by users.Slots
optTx:- Object of class
"numeric."
Estimated optimal treatment for training data.
Coded as +1.0/-1.0 independent of
notation provided in original data. estVal:- Object of class
"numeric."
Estimated value of regime for training data. regime:- Object of class
"formula."
Formula description of covariates used in
kernel. crossValidation:- Object of class
"CVInfoOrNULL."
Cross-validation results for training data. optim:- Object of class
"OWLOptim."
Optimization results. decisionFunc:- Object of class
"numeric."
Estimated decision function for origina data.
Methods
- cvInfo
signature(object = "BOWLBasic"):
Retrieve cross-validation matrix.- optimObj
signature(object = "BOWLBasic"):
Retrieve optimization results. - .predictOptimalTx
signature(x = "BOWLBasic", newdata = "missing"):
Retrieve estimated optimal treatment for training data.
Method is not exported.- .predictOptimalTx
signature(x = "BOWLBasic", newdata = "data.frame"):
Estimate optimal treatment for newdata. Method is not exported.- print
signature(x = "BOWLBasic"):
Print key results of method. - regimeCoef
signature(object = "BOWLBasic"):
Retrieve parameter estimates for decision function. - show
signature(object = "BOWLBasic"):
Show key results of method. - summary
signature(object = "BOWLBasic"):
Retrieve key summary information of method.