TMLE (Targeted Maximum Likelihood Estimation)

Use the synthetic data setup first; it defines data and the columns used below. These are advanced direct R examples.

Target outcome-model estimates using propensity information. Binary-outcome contrasts are reported on the risk-difference scale.

tmle_result <- causal_tmle(data, "Immunotherapy", "Response",
                           confounders = c("Age", "Sex", "Stage", "ECOG"),
                           outcome_type = "binary")

Read the result

Optional TMLE dependencies must be installed for eligible calls. Non-survival outputs are RiskDiff or MeanDiff. Survival and continuous-policy branches have separate support limits; a runtime support label is not a current validation result.

See effect scales and reporting, diagnostics and the current support reference.