Skip to contents

Sandwich variance for the AIPW-lambda estimator

var_beta_aipw_lambda.Rd

Computes a sandwich variance estimator for the AIPW-lambda estimator under outcome-dependent right-censoring of a covariate, using the closed-form augmentation and accounting for estimation of the censoring model \(C | (Y, Z)\) via a Gumbel/Weibull AFT model.

Usage

var_beta_aipw_lambda(data_yXZ, mytheta)

Arguments

data_yXZ

Data frame containing at least:

  • y: outcome,

  • A: auxiliary covariate,

  • AW: A - X (used in the outcome model),

  • W: observed W = min(X, C),

  • D: indicator I(X <= C),

  • Z: covariate in the outcome and censoring models,

  • myp_ywz: weights \(\pi(Y, W, Z)\) from AFT model (for IPW).

mytheta

Numeric vector c(beta0, beta1, beta2, psi) corresponding to the AIPW-lambda point estimates from estimate_beta_aipw_lambda_close().

Value

A list with components:

beta_est

The input parameter vector mytheta.

se_est

Sandwich standard errors for \(\beta_0, \beta_1, \beta_2\).

Details

This implementation assumes the outcome model is y ~ AW + Z, i.e., three regression coefficients \((\beta_0, \beta_1, \beta_2)\) plus a residual standard deviation \(\psi\).