polle1.6.2 package

Policy Learning

c_model

c_model class object

conditional

Conditional Policy Evaluation

control_blip

Control arguments for doubly robust blip-learning

control_drql

Control arguments for doubly robust Q-learning

control_earl

Control arguments for Efficient Augmentation and Relaxation Learning

control_owl

Control arguments for Outcome Weighted Learning

control_ptl

Control arguments for Policy Tree Learning

control_rwl

Control arguments for Residual Weighted Learning

copy_policy_data

Copy Policy Data Object

fit_c_functions

Fit Censoring Functions

fit_g_functions

Fit g-functions

g_model

g_model class object

get_action_set

Get Action Set

get_actions

Get Actions

get_event

Get event indicators

get_g_functions

Get g-functions

get_history_names

Get history variable names

get_history

Get History Object

get_id_stage

Get IDs and Stages

get_id

Get IDs

get_K

Get Maximal Stages

get_n

Get Number of Observations

get_policy_actions

Get Policy Actions

get_policy_functions

Get Policy Functions

get_policy_object

Get Policy Object

get_policy

Get Policy

get_q_functions

Get Q-functions

get_stage_action_sets

Get Stage Action Sets

get_utility

Get the Utility

nuisance_functions

Nuisance Functions

partial

Trim Number of Stages

plot.policy_data

Plot policy data for given policies

plot.policy_eval

Plot histogram of the influence curve for a policy_eval object

policy_data

Create Policy Data Object

policy_def

Define Policy

policy_eval_online

Online/Sequential Policy Evaluation

policy_eval

Policy Evaluation

policy_learn

Create Policy Learner

policy

Policy-class

polle-package

polle: Policy Learning

predict.nuisance_functions

Predict g-functions and Q-functions

q_model

q_model class object

reexports

Objects exported from other packages

sim_multi_stage

Simulate Multi-Stage Data

sim_single_stage_multi_actions

Simulate Single-Stage Multi-Action Data

sim_single_stage

Simulate Single-Stage Data

sim_two_stage_multi_actions

Simulate Two-Stage Multi-Action Data

sim_two_stage

Simulate Two-Stage Data

subset_id

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.

  • Maintainer: Andreas Nordland
  • License: Apache License (>= 2)
  • Last published: 2025-12-04