V_p: Oracular approximation of value function

In PLUCR: Policy Learning Under Constraint

Oracular approximation of value function

Description

Computes the expected outcome under a policy determined by the previously optimized psi(X). The policy assigns treatment probabilistically based on sigma_beta(psi(X)), and the expected outcome is calculated using counterfactual outcomes.

Usage

V_p(
  psi,
  beta = 0.05,
  centered = FALSE,
  alpha = 0.1,
  B = 1e+06,
  ncov = 10L,
  scenario_mu = c("Linear", "Threshold", "Mix", "Linear2", "Null", "Realistic"),
  scenario_nu = c("Linear", "Threshold", "Mix", "Satisfied", "Realistic"),
  seed = NA
)

Arguments

psi	A function that takes an input X and returns a numeric vector with values in the range ⁠[-1,1]⁠.
beta	A non-negative numeric scalar controlling the sharpness of the probability function (0.05 by default).
centered	A logical value indicating whether to apply centering in sigma_beta (FALSE by default).
alpha	A numeric scalar representing the constraint tolerance (in ⁠[0,1/2]⁠, 0.1 by default).
B	Integer, number of Monte Carlo repetitions (1e4 by default).
ncov	Number of baseline covariates (at least 2L and 10L by default).
scenario_mu	String indicating the type of scenario for delta_Mu ("Linear", "Threshold", "Mix").
scenario_nu	String indicating the type of scenario for delta_Nu ("Linear", "Threshold", "Mix").
seed	Integer or NA (NA by default).

Value

A numeric scalar representing the expected primary outcome under the policy.

PLUCR documentation built on March 30, 2026, 5:08 p.m.