Policy Learning
c_model class object
Conditional Policy Evaluation
Control arguments for doubly robust blip-learning
Control arguments for doubly robust Q-learning
Control arguments for Efficient Augmentation and Relaxation Learning
Control arguments for Outcome Weighted Learning
Control arguments for Policy Tree Learning
Control arguments for Residual Weighted Learning
Copy Policy Data Object
Fit Censoring Functions
Fit g-functions
g_model class object
Get Action Set
Get Actions
Get event indicators
Get g-functions
Get history variable names
Get History Object
Get IDs and Stages
Get IDs
Get Maximal Stages
Get Number of Observations
Get Policy Actions
Get Policy Functions
Get Policy Object
Get Policy
Get Q-functions
Get Stage Action Sets
Get the Utility
Nuisance Functions
Trim Number of Stages
Plot policy data for given policies
Plot histogram of the influence curve for a policy_eval object
Create Policy Data Object
Define Policy
Online/Sequential Policy Evaluation
Policy Evaluation
Create Policy Learner
Policy-class
polle: Policy Learning
Predict g-functions and Q-functions
q_model class object
Objects exported from other packages
Simulate Multi-Stage Data
Simulate Single-Stage Multi-Action Data
Simulate Single-Stage Data
Simulate Two-Stage Multi-Action Data
Simulate Two-Stage Data
Subset Policy Data on ID
Package for learning and evaluating (subgroup) policies via doubly robust loss functions. Policy learning methods include doubly robust blip/conditional average treatment effect learning and sequential policy tree learning. Methods for (subgroup) policy evaluation include doubly robust cross-fitting and online estimation/sequential validation. See Nordland and Holst (2022) <doi:10.48550/arXiv.2212.02335> for documentation and references.