as.classCoMiRe

A constructor for the <code>classCoMiRe</code> class. The class <code>classCoMiRe</code> is a named list containing
the output of the posterior estimation of CoMiRe model implemented in <code>comire.gibbs</code>

Posterior inference under the convex mixture regression (CoMiRe) models introduced by Canale, Durante, and Dunson (2018) <doi:10.1111/biom.12917>.

Antonio Canale

CoMiRe

Convex Mixture Regression

as.classCoMiRe function

<dl><dt>call</dt>
<dd>a formula for <code>comire.gibbs</code>.</dd>
<dt>out</dt>
<dd>an output of <code>comire.gibbs</code>.</dd>
<dt>z</dt>
<dd>optional numeric vector or matrix for the confounding covariates.</dd>
<dt>z.val</dt>
<dd>optional numeric vector containing a fixed value of interest for each of the confounding covariates to be used for the plots. Default value is <code>mean(z)</code> for numeric covariates or the mode for factorial covariates.</dd>
<dt>f0, f1</dt>
<dd>optional matrices containing simulated values of the mixture densities at low and high dose exposure; default values are simulated with <code>comire.gibbs</code>. It is possible to change these for differente fixed values of <code>z</code>: see <code>predict_new_z</code> function.</dd>
<dt>nrep</dt>
<dd>integer giving the total number of iterations used in <code>comire.gibbs</code>.</dd>
<dt>nb</dt>
<dd>integer giving the number of burn-in iterations used in <code>comire.gibbs</code>.</dd>
<dt>bin</dt>
<dd>logical. It is <code>TRUE</code> if y is drawn for a binomial distribution.</dd>
<dt>univariate</dt>
<dd>logical. It is <code>TRUE</code> if the model is univariate.</dd></dl>

Arguments

Author

classCoMiRe class constructor — as.classCoMiRe

<dl>

<dt>call</dt>
<dd>a formula for <code>comire.gibbs</code>.</dd>


<dt>out</dt>
<dd>an output of <code>comire.gibbs</code>.</dd>


<dt>z</dt>
<dd>optional numeric vector or matrix for the confounding covariates.</dd>


<dt>z.val</dt>
<dd>optional numeric vector containing a fixed value of interest for each of the confounding covariates to be used for the plots. Default value is <code>mean(z)</code> for numeric covariates or the mode for factorial covariates.</dd>


<dt>f0, f1</dt>
<dd>optional matrices containing simulated values of the mixture densities at low and high dose exposure; default values are simulated with <code>comire.gibbs</code>. It is possible to change these for differente fixed values of <code>z</code>: see <code>predict_new_z</code> function.</dd>


<dt>nrep</dt>
<dd>integer giving the total number of iterations used in <code>comire.gibbs</code>.</dd>


<dt>nb</dt>
<dd>integer giving the number of burn-in iterations used in <code>comire.gibbs</code>.</dd>


<dt>bin</dt>
<dd>logical. It is <code>TRUE</code> if y is drawn for a binomial distribution.</dd>


<dt>univariate</dt>
<dd>logical. It is <code>TRUE</code> if the model is univariate.</dd>

</dl>

as.classCoMiRe: classCoMiRe class constructor

Description

Usage

Arguments

Author