High-dimensional sparse partially linear DGP (v1.4.0) — dgp_synth_hd_sparse_plm

Version bump that increases dimensionality and hardens collinearity: \(X \in \mathbb{R}^{100}\) with Toeplitz correlation \(\Sigma_{ij} = 0.95^{|i-j|}\).

Usage

dgp_synth_hd_sparse_plm_v140(
  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 synthetic DGP list with df, true_att, true_qst, and meta.

Details

The outcome regression remains sparse in the first 5 covariates, but the propensity signal is strengthened (still sparse) to push standard propensity modeling into a separation-like regime when \(N/P\) is small.