cde_aipw

Initializes the specification of a CDE estimator based on an
augmented inverse probability weighting approach.

A set of functions to estimate the controlled direct effect of treatment fixing a potential mediator to a specific value. Implements the sequential g-estimation estimator described in Vansteelandt (2009) <doi:10.1097/EDE.0b013e3181b6f4c9> and Acharya, Blackwell, and Sen (2016) <doi:10.1017/S0003055416000216> and the telescope matching estimator described in Blackwell and Strezhnev (2020) <doi:10.1111/rssa.12759>.

Matthew Blackwell

DirectEffects

Estimating Controlled Direct Effects for Explaining Causal
Findings

Avidit Acharya

Maya Sen

Shiro Kuriwaki

Jacob Brown

Anton Strezhnev

cde_aipw function

<dl><dt>trim</dt>
<dd>A vector of length 2 indicating what quantiles of the
propensity scores should be trimmed. By default this is <code>c(0.01, 0.99)</code> meaning that the top and bottom 1% of propensity scores are
trunctated to these quantiles. If NULL, no trimming occurs.</dd>
<dt>aipw_blip</dt>
<dd>If <code>TRUE</code> (the default), augmented inverse probability weighting
estimators will be used to estimate intermediate outcome
regressions (blip functions).</dd></dl>

Arguments

Initialize an AIPW CDE estimator — cde_aipw

<dl>

<dt>trim</dt>
<dd>A vector of length 2 indicating what quantiles of the
propensity scores should be trimmed. By default this is <code>c(0.01, 0.99)</code> meaning that the top and bottom 1% of propensity scores are
trunctated to these quantiles. If NULL, no trimming occurs.</dd>


<dt>aipw_blip</dt>
<dd>If <code>TRUE</code> (the default), augmented inverse probability weighting
estimators will be used to estimate intermediate outcome
regressions (blip functions).</dd>

</dl>

cde_aipw: Initialize an AIPW CDE estimator

Description

Usage

Arguments