Wraps a balancer to automatically double the sample size N if the original
limit is exceeded.
Public fields
balancerThe underlying balancer object.
clsThe class generator (e.g., BWD or MultiBWD).
cls_nameThe string name of the class (for serialization).
Active bindings
definitionDictionary of definition parameters.
stateDictionary of current state.
Methods
Method new()
Initialize the Online wrapper.
Arguments
clsThe class generator (e.g., BWD) OR its character string name.
...Arguments passed to the balancer's initialize method.
Method assign_next()
Assign treatment to the next point, expanding N if necessary.
Arguments
xCovariate profile vector.
Returns
Treatment assignment.
Method assign_all()
Delegate assign_all to inner balancer.
Arguments
XMatrix of covariate profiles.
Method update_state()
Delegate update_state to inner balancer.
Method reset()
Delegate reset to inner balancer.
Method process_x()
Delegate process_x to inner balancer.
Method update_imbalance()
Delegate update_imbalance to inner balancer.
Usage
Online$update_imbalance(x, a)
Arguments
xProcessed covariate vector.
aAssignment.
Method update_path()
Delegate update_path to inner balancer.
Arguments
xCovariate vector.
aAssignment.
Method clone()
The objects of this class are cloneable with this method.
Usage
Online$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.