Mildly tilted propensity synthetic DGP for CausalStress (v1.3.0) — dgp_synth_tilt_mild

Implements the synth_tilt_mild design from the DGP registry. Uses the same outcome and noise as synth_baseline but a mildly tilted propensity to induce covariate shift.

Usage

dgp_synth_tilt_mild_v130(
  n,
  seed = NULL,
  include_truth = TRUE,
  oracle_only = FALSE
)

Arguments

n: Integer, number of observations.
seed: Optional seed for reproducibility (passed to cs_set_rng()).
include_truth: Logical; if TRUE, include oracle truth tables where supported.
oracle_only: Logical; if TRUE, return only columns needed for oracle truth generation where supported.

Value

A list with df, true_att, true_qst, and meta following the synthetic DGP contract.

Details

Goal: Test basic propensity reweighting under mild covariate shift.

Structure: Outcomes/noise match synth_baseline.

Covariates: \(X \sim \mathcal{N}(0, I_5)\) (global default).

Propensity: \(p(X) = \mathrm{plogis}(0.45 X_1 - 0.3 X_2 - 0.25 X_4)\).

Clipping: None in the DGP; estimators commonly clip to \([0.05, 0.95]\).