This function implements augmented inverse probability weighted (IPW) estimation of average treatment effects (ATEs), provided both fitted propensity scores and fitted values from outcome regression.
ate.aipw(y, tr, mfp, mfo, off = NULL)An \(n\) x \(1\) vector of observed outcomes.
An \(n\) x \(1\) vector of treatment indicators (=1 if treated or 0 if untreated).
An \(n\) x \(2\) matrix of fitted propensity scores for untreated (first column) and treated (second column).
An \(n\) x \(2\) matrix of fitted values from outcome regression, for untreated (first column) and treated (second column).
A \(2\) x \(1\) vector of offset values (e.g., the true values in simulations) used to calculate the z-statistics.
A \(2\) x \(1\) vector of direct IPW estimates of 1.
A \(2\) x \(1\) vector of ratio IPW estimates of means.
A \(2\) x \(1\) vector of outcome regression estimates of means.
A \(2\) x \(1\) vector of augmented IPW estimates of means.
The estimated variances associated with the augmented IPW estimates of means.
The z-statistics for the augmented IPW estimates of means, compared to off.
The augmented IPW estimate of ATE.
The estimated variance associated with the augmented IPW estimate of ATE.
The z-statistic for the augmented IPW estimate of ATE.
Tan, Z. (2020a) Regularized calibrated estimation of propensity scores with model misspecification and high-dimensional data, Biometrika, 107, 137<U+2013>158.
Tan, Z. (2020b) Model-assisted inference for treatment effects using regularized calibrated estimation with high-dimensional data, Annals of Statistics, 48, 811<U+2013>837.