Skip to contents

PGC binary estimator: proxy G-component correction with logistic / LPM

Source: R/bin_estimators.R
fit_pgc_bin.Rd

Three-step bridge-function estimator with a binary outcome stage:

  1. Residualise X on G -> X_resid (OLS).

  2. Bridge X_resid on the FULL W matrix -> W_hat (OLS).

  3. Regress the binary outcome on X + W_hat via logistic regression (log-OR) or a linear probability model (risk difference).

Usage

fit_pgc_bin(y, X, g, W, covars = NULL, effect_scale = c("logor", "riskdiff"))

Arguments

y

Numeric 0/1 outcome vector (length n).

X

Numeric exposure vector (length n).

g

Numeric instrument vector (length n).

W

Numeric NC matrix (n x q) or vector (length n).

covars

Optional data frame of covariates (n rows).

effect_scale

Character: "logor" or "riskdiff".

Value

Named list: beta, se, pvalue.

Examples

set.seed(1)
dat <- generate_toy_data(n = 200, outcome_type = "binary", seed = 1)
fit_pgc_bin(dat$y_bin, dat$X, dat$G[, 1], dat$W)
#> $beta
#> [1] 0.2603184
#> 
#> $se
#> [1] 0.1982229
#> 
#> $pvalue
#> [1] 0.189095
#>