qscore_impl

Generates the quasi-score statistic for a generalised linear mixed model

Frequentist statistical inference for cluster randomised trials with
multiple outcomes that controls the family-wise error rate and provides nominal
coverage of confidence sets. A full description of the methods can be found in Watson et al. (2023) <doi:10.1002/sim.9831>.

Sam Watson

crctStepdown

Univariate Analysis of Cluster Trials with Multiple Outcomes

qscore_impl function

<dl><dt>resids</dt>
<dd>A numeric vector of generalised residuals</dd>
<dt>tr</dt>
<dd>A numeric vector of 1s (treatment group) and -1s (control group)</dd>
<dt>xb</dt>
<dd>A numeric vector of fitted linear predictors</dd>
<dt>invS</dt>
<dd>A matrix. If using the weighted statistic then it should be the inverse covariance matrix of the observations</dd>
<dt>family2</dt>
<dd>A string naming the link function</dd>
<dt>Z</dt>
<dd>A matrix with columns indicating cluster membership</dd>
<dt>weight</dt>
<dd>Logical value indicating whether to use the weighted statistic (TRUE) or the unweighted statistic (FALSE)</dd></dl>

Arguments

The quasi-score statistic for a generalised linear mixed model — qscore_impl

<dl>

<dt>resids</dt>
<dd>A numeric vector of generalised residuals</dd>


<dt>tr</dt>
<dd>A numeric vector of 1s (treatment group) and -1s (control group)</dd>


<dt>xb</dt>
<dd>A numeric vector of fitted linear predictors</dd>


<dt>invS</dt>
<dd>A matrix. If using the weighted statistic then it should be the inverse covariance matrix of the observations</dd>


<dt>family2</dt>
<dd>A string naming the link function</dd>


<dt>Z</dt>
<dd>A matrix with columns indicating cluster membership</dd>


<dt>weight</dt>
<dd>Logical value indicating whether to use the weighted statistic (TRUE) or the unweighted statistic (FALSE)</dd>

</dl>

qscore_impl: The quasi-score statistic for a generalised linear mixed model

Description

Usage

Value

Arguments