Multi-Method Comparison

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

Compare estimates for compatible targets and effect scales. ORs, risk differences, HRs and survival contrasts are not interchangeable.

compare_result <- causal_compare_all(
  data,
  treatment = "Immunotherapy",
  outcome = "Response",
  confounders = c("Age", "Sex", "Stage", "ECOG"),
  outcome_type = "binary",
  methods = c("GLM", "GComp", "AIPW", "IPW", "PSM", "CausalForest"),
  boot_R = 500
)

# Method comparison table
compare_methods(compare_result)

# Forest plot
iobr_causal_plot(compare_result, type = "forest")

# Method consistency plot
iobr_causal_plot(compare_result, type = "consistency")

Read the result

Keep requested, eligible, run, skipped and failed methods visible. A shared direction does not make differently scaled estimates numerically comparable. Several methods can share assumptions or backends; their agreement is not independent causal confirmation.

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