ate.aipw {RCAL}

Augmented inverse probability weighted estimation of population means

Description

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.

Usage

ate.aipw(y, tr, mfp, mfo, off = NULL)

Arguments

y

An n x 1 vector of observed outcomes.

tr

An n x 1 vector of treatment indicators (=1 if treated or 0 if untreated).

mfp

An n x 2 matrix of fitted propensity scores for untreated (first column) and treated (second column).

mfo

An n x 2 matrix of fitted values from outcome regression, for untreated (first column) and treated (second column).

off

A 2 x 1 vector of offset values (e.g., the true values in simulations) used to calculate the z-statistics.

Value

one

A 2 x 1 vector of direct IPW estimates of 1.

ipw

A 2 x 1 vector of ratio IPW estimates of means.

or

A 2 x 1 vector of outcome regression estimates of means.

est

A 2 x 1 vector of augmented IPW estimates of means.

var

The estimated variances associated with the augmented IPW estimates of means.

ze

The z-statistics for the augmented IPW estimates of means, compared to off.

diff

The augmented IPW estimate of ATE.

diff.var

The estimated variance associated with the augmented IPW estimate of ATE.

diff.ze

The z-statistic for the augmented IPW estimate of ATE.

References

Tan, Z. (2020a) Regularized calibrated estimation of propensity scores with model misspecification and high-dimensional data, Biometrika, 107, 137–158.

Tan, Z. (2020b) Model-assisted inference for treatment effects using regularized calibrated estimation with high-dimensional data, Annals of Statistics, 48, 811–837.