estimate

Fit a CDE estimator with the engines specified in the <code>model_spec</code>
object.

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

estimate function

<dl><dt>object</dt>
<dd>A <code>cde_estimator</code> object that has already been passed
to at least one call to <code>set_treatment</code>.</dd>
<dt>formula</dt>
<dd>A formula object with describing the outcome of
interest on the left-hand side and the treatment variables the user
wants to estimate effects for (which might be a subset of the
treatment variables specified).</dd>
<dt>data</dt>
<dd>A <code>data.frame</code> containing all variables, including
treatment variables and covariates specified.</dd>
<dt>subset</dt>
<dd>Anan optional vector specifying a subset of
observations to be used in the fitting process.</dd>
<dt>crossfit</dt>
<dd>A logical indicator for if cross-fitting should be
used in estimating the effects.</dd>
<dt>n_folds</dt>
<dd>The number of folds to use within a given instance
of the cross-fitting algorithm.</dd>
<dt>n_splits</dt>
<dd>The number of times the cross-fitting procedure
should be repeated. Overall estimates use the median value of these
repeated estimates.</dd></dl>

Arguments

Fit a specified CDE estimator — estimate

<dl>

<dt>object</dt>
<dd>A <code>cde_estimator</code> object that has already been passed
to at least one call to <code>set_treatment</code>.</dd>


<dt>formula</dt>
<dd>A formula object with describing the outcome of
interest on the left-hand side and the treatment variables the user
wants to estimate effects for (which might be a subset of the
treatment variables specified).</dd>


<dt>data</dt>
<dd>A <code>data.frame</code> containing all variables, including
treatment variables and covariates specified.</dd>


<dt>subset</dt>
<dd>Anan optional vector specifying a subset of
observations to be used in the fitting process.</dd>


<dt>crossfit</dt>
<dd>A logical indicator for if cross-fitting should be
used in estimating the effects.</dd>


<dt>n_folds</dt>
<dd>The number of folds to use within a given instance
of the cross-fitting algorithm.</dd>


<dt>n_splits</dt>
<dd>The number of times the cross-fitting procedure
should be repeated. Overall estimates use the median value of these
repeated estimates.</dd>

</dl>

estimate: Fit a specified CDE estimator

Description

Usage

Arguments