A fitted uplift / next-best-action model

The object returned by fit_uplift(): a joint gmm_fit over (outcome, treatment, covariates) together with the column roles, the identification assumption it will be read under, the outcome type, and the training sample (retained so that resampling standard errors and overlap diagnostics are available). The decision verbs dispatch on this class.

Usage

uplift_model(
  fit = NULL,
  roles = list(),
  assume = "ignorability",
  outcome_type = "continuous",
  data = NULL,
  n_train = integer(0),
  treatment_levels = c(0, 1),
  name = "uplift_model",
  metadata = list()
)

Arguments

fit

The joint gmm_fit over the stacked (outcome, treatment, covariates) coordinates, in that column order.

roles

A list with integer indices outcome, treatment, covariate and the matching outcome_name, treatment_name, covariate_names.

assume

One of "ignorability" (the default uplift assumption) or "latent_confounder" (the do-operator reading; always flagged).

outcome_type

One of "continuous", "binary", "count".

data

The n by p numeric training matrix cbind(outcome, treatment, covariates).

n_train

Integer scalar – the training sample size.

treatment_levels

Numeric length-2 vector c(t0, t1) – the control and treated values used at fit time (the observed treatment arms).

name

Human-readable name.

metadata

Optional list of descriptors (e.g. the K-selection trace).

Value

An S7 object of class uplift_model.

See also

Other decision: fit_uplift(), gmm_cf_mean(), gmm_cf_tail_prob(), gmm_cf_variance(), gmm_counterfactual(), gmm_intervene(), proxy_cate(), proxy_confounding_gap(), proxy_decide(), proxy_identification_report(), proxy_overlap(), proxy_policy_value(), proxy_predict(), proxy_regime_segments(), proxy_retrospective_uplift(), proxy_uplift(), uplift_identification()