gsc_inference: Non-parametric Inference for GSC (Xu 2017)

In coresynth: Fast and Unified Synthetic Control Methods

Non-parametric Inference for GSC (Xu 2017)

Description

Estimates SE and confidence intervals for the ATT via non-parametric cluster bootstrap or jackknife over control units. Works for both sharp and staggered GSC fits. For staggered fits, bootstrap resamples each cohort's control pool independently, and jackknife uses a per-cohort LOO with delta-method variance aggregation.

Usage

gsc_inference(
  fit,
  method = c("bootstrap", "jackknife", "jackknife_global"),
  n_boot = 499L,
  level = 0.95,
  alternative = c("two.sided", "greater", "less"),
  seed = NULL
)

Arguments

fit	A coresynth object from scm_fit() with method = "gsc".
method	"bootstrap" (default) or "jackknife".
n_boot	Number of bootstrap replications (default 499L; ignored for jackknife).
level	Confidence level (default 0.95).
alternative	"two.sided" (default), "greater", or "less".
seed	RNG seed for reproducibility (default NULL).

Details

Note: gsc_boot() performs a parametric bootstrap under H0 (hypothesis testing). gsc_inference() provides non-parametric SE and CIs suitable for inference about the ATT magnitude.

Value

A list of class coresynth_inference.

coresynth documentation built on Aug. 28, 2026, 1:06 a.m.