posterior_predict.clmstan

Draws from the posterior predictive distribution. For each posterior
sample, a predicted category is sampled from the categorical distribution
with the predicted probabilities.

Fits cumulative link models (CLMs) for ordinal categorical data
using 'CmdStanR'. Supports various link functions including logit, probit,
cloglog, loglog, cauchit, and flexible parametric links such as
Generalized Extreme Value (GEV), Asymmetric Exponential Power (AEP),
and Symmetric Power. Models are pre-compiled using the 'instantiate'
package for fast execution without runtime compilation. Methods are
described in Agresti (2010, ISBN:978-0-470-08289-8),
Wang and Dey (2011) <doi:10.1007/s10651-010-0154-8>, and
Naranjo, Perez, and Martin (2015) <doi:10.1007/s11222-014-9449-1>.

Tomotaka Momozaki

clmstan

Cumulative Link Models with 'CmdStanR'

posterior_predict.clmstan function

<dl><dt>object</dt>
<dd>A <code>clmstan</code> object returned by <code>clm_stan()</code>.</dd>
<dt>newdata</dt>
<dd>Optional data frame for prediction. If <code>NULL</code> (default),
predictions are made for the original training data.</dd>
<dt>ndraws</dt>
<dd>Number of posterior draws to use. If <code>NULL</code> (default),
all available draws are used.</dd>
<dt>...</dt>
<dd>Additional arguments (currently ignored).</dd></dl>

Arguments

Posterior predictive distribution for clmstan objects — posterior_predict.clmstan

<dl>

<dt>object</dt>
<dd>A <code>clmstan</code> object returned by <code>clm_stan()</code>.</dd>


<dt>newdata</dt>
<dd>Optional data frame for prediction. If <code>NULL</code> (default),
predictions are made for the original training data.</dd>


<dt>ndraws</dt>
<dd>Number of posterior draws to use. If <code>NULL</code> (default),
all available draws are used.</dd>


<dt>...</dt>
<dd>Additional arguments (currently ignored).</dd>

</dl>

posterior_predict.clmstan: Posterior predictive distribution for clmstan objects

Description

Usage

Value

Arguments

See Also