fitting_specification

The <code>brma</code> family of functions uses the following arguments to specify the
MCMC sampling and fitting settings.

A framework for Bayesian meta-analysis, including model estimation,
prior specification, model comparison, prediction, summaries, visualizations,
and diagnostics. The package fits single and model-averaged meta-analytic,
meta-regression, multilevel, publication bias adjusted, and generalized linear
mixed models The model-averaged meta-analytic models combine competing models
based on their predictive performance, weight inference by posterior model probabilities,
and test model components using Bayes factors (e.g., effect vs. no effect;
Bartoš et al., 2022, <doi:10.1002/jrsm.1594>; Maier, Bartoš & Wagenmakers, 2022,
<doi:10.1037/met0000405>; Bartoš et al., 2025, <doi:10.1037/met0000737>).
Users can specify flexible prior distributions for effect sizes, heterogeneity,
publication bias (including selection models and PET-PEESE), and moderators.

Franti<c5><a1>ek Barto<c5><a1>

RoBMA

Robust Bayesian Meta-Analyses

František Bartoš

Maximilian Maier

Eric-Jan Wagenmakers

Joris Goosen

Matthew Denwood

Martyn Plummer

fitting_specification function

<dl><dt>sample</dt>
<dd>numeric. Number of MCMC samples to save. Defaults to <code>5000</code>.</dd>
<dt>burnin</dt>
<dd>numeric. Number of burn-in iterations. Defaults to <code>2000</code>.</dd>
<dt>adapt</dt>
<dd>numeric. Number of adaptation iterations. Defaults to <code>500</code>.</dd>
<dt>chains</dt>
<dd>numeric. Number of MCMC chains. Defaults to <code>3</code>.</dd>
<dt>thin</dt>
<dd>numeric. Thinning interval. Defaults to <code>1</code>.</dd>
<dt>parallel</dt>
<dd>logical. Whether to run MCMC chains in parallel. Defaults to <code>FALSE</code>.</dd>
<dt>autofit</dt>
<dd>logical. Whether to automatically extend the MCMC chains if convergence is not met.
Defaults to <code>FALSE</code>.</dd>
<dt>autofit_control</dt>
<dd>list of autofit control settings. See <code>set_autofit_control()</code> for details.</dd>
<dt>convergence_checks</dt>
<dd>list of convergence check settings. See <code>set_convergence_checks()</code> for details.</dd>
<dt>seed</dt>
<dd>numeric. Random seed for reproducibility. Defaults to <code>NULL</code>.</dd>
<dt>silent</dt>
<dd>logical. Whether to suppress output. Constructors with no
explicit default use <code>RoBMA.get_option("silent")</code> when <code>silent</code> is omitted.
Model-averaging wrappers default to <code>TRUE</code> unless explicitly changed.</dd>
<dt>...</dt>
<dd>additional advanced arguments. Fitting functions reject unused
arguments; currently recognized internal arguments include <code>only_data</code>,
<code>only_priors</code>, <code>is_JASP</code>, and <code>is_JASP_prefix</code>.</dd></dl>

Arguments

Fitting specification — fitting_specification

<dl>

<dt>sample</dt>
<dd>numeric. Number of MCMC samples to save. Defaults to <code>5000</code>.</dd>


<dt>burnin</dt>
<dd>numeric. Number of burn-in iterations. Defaults to <code>2000</code>.</dd>


<dt>adapt</dt>
<dd>numeric. Number of adaptation iterations. Defaults to <code>500</code>.</dd>


<dt>chains</dt>
<dd>numeric. Number of MCMC chains. Defaults to <code>3</code>.</dd>


<dt>thin</dt>
<dd>numeric. Thinning interval. Defaults to <code>1</code>.</dd>


<dt>parallel</dt>
<dd>logical. Whether to run MCMC chains in parallel. Defaults to <code>FALSE</code>.</dd>


<dt>autofit</dt>
<dd>logical. Whether to automatically extend the MCMC chains if convergence is not met.
Defaults to <code>FALSE</code>.</dd>


<dt>autofit_control</dt>
<dd>list of autofit control settings. See <code>set_autofit_control()</code> for details.</dd>


<dt>convergence_checks</dt>
<dd>list of convergence check settings. See <code>set_convergence_checks()</code> for details.</dd>


<dt>seed</dt>
<dd>numeric. Random seed for reproducibility. Defaults to <code>NULL</code>.</dd>


<dt>silent</dt>
<dd>logical. Whether to suppress output. Constructors with no
explicit default use <code>RoBMA.get_option("silent")</code> when <code>silent</code> is omitted.
Model-averaging wrappers default to <code>TRUE</code> unless explicitly changed.</dd>


<dt>...</dt>
<dd>additional advanced arguments. Fitting functions reject unused
arguments; currently recognized internal arguments include <code>only_data</code>,
<code>only_priors</code>, <code>is_JASP</code>, and <code>is_JASP_prefix</code>.</dd>

</dl>

fitting_specification: Fitting specification

Description

Arguments

See Also