get_tau_vector

Creates a sequence of tau values spanning a confidence interval around
the estimated treatment effect, with oversampling near the point estimate.

internal

Implements methods developed by Ding, Feller, and Miratrix (2016) <doi:10.1111/rssb.12124> <doi:10.48550/arXiv.1412.5000>,
and Ding, Feller, and Miratrix (2018) <doi:10.1080/01621459.2017.1407322> <doi:10.48550/arXiv.1605.06566>
for testing whether there is unexplained variation in treatment effects across observations, and for characterizing
the extent of the explained and unexplained variation in treatment effects. The package includes wrapper functions
implementing the proposed methods, as well as helper functions for analyzing and visualizing the results of the test.

Ben Fifield

hettx

Fisherian and Neymanian Methods for Detecting and Measuring
Treatment Effect Variation

Peng Ding

Avi Feller

Luke Miratrix

get_tau_vector function

<dl><dt>Y</dt>
<dd>Outcome vector.</dd>
<dt>Z</dt>
<dd>Treatment assignment vector (0/1).</dd>
<dt>X</dt>
<dd>Optional covariate matrix for regression-adjusted estimation.</dd>
<dt>gamma</dt>
<dd>Width of the confidence interval. Default is 0.001.</dd>
<dt>grid.size</dt>
<dd>Number of grid points. Default is 21.</dd>
<dt>grid.gamma</dt>
<dd>Controls oversampling density near the estimate. Default is 100*gamma.</dd></dl>

Arguments

Generate a grid of treatment effect values to test — get_tau_vector

<dl>

<dt>Y</dt>
<dd>Outcome vector.</dd>


<dt>Z</dt>
<dd>Treatment assignment vector (0/1).</dd>


<dt>X</dt>
<dd>Optional covariate matrix for regression-adjusted estimation.</dd>


<dt>gamma</dt>
<dd>Width of the confidence interval. Default is 0.001.</dd>


<dt>grid.size</dt>
<dd>Number of grid points. Default is 21.</dd>


<dt>grid.gamma</dt>
<dd>Controls oversampling density near the estimate. Default is 100*gamma.</dd>

</dl>

get_tau_vector: Generate a grid of treatment effect values to test

Description

Usage

Value

Arguments