pr_MLiG

Create an object representing a Multivariate Log inverse Gamma (MLiG) prior distribution

Fit multi-level models with possibly correlated
random effects using Markov Chain Monte Carlo simulation.
Such models allow smoothing over space and time and are useful in,
for example, small area estimation.

Harm Jan Boonstra

mcmcsae

Markov Chain Monte Carlo Small Area Estimation

Grzegorz Baltissen

pr_MLiG function

<dl><dt>mean</dt>
<dd>scalar or vector parameter for the mean in the large
<code>a</code> limit, when the distribution approaches a normal distribution.</dd>
<dt>precision</dt>
<dd>scalar or vector parameter for the precision in the
large <code>a</code> limit, when the distribution approaches a normal
distribution.</dd>
<dt>labels</dt>
<dd>optional character vector with coefficient labels. If specified,
it should have the same length as at least one of <code>mean</code> and <code>precision</code>,
and in that case the MLiG prior with these parameters is assigned to these coefficients,
while any coefficients not present in labels will be assigned a non-informative
prior with mean 0 and precision 0.</dd>
<dt>a</dt>
<dd>scalar parameter that controls how close the prior is to independent
normal priors with <code>mean</code> and <code>precision</code> parameters. The larger
this value (default is 1000), the closer.</dd></dl>

Arguments

Create an object representing a Multivariate Log inverse Gamma (MLiG) prior distribution — pr_MLiG

<dl>

<dt>mean</dt>
<dd>scalar or vector parameter for the mean in the large
<code>a</code> limit, when the distribution approaches a normal distribution.</dd>


<dt>precision</dt>
<dd>scalar or vector parameter for the precision in the
large <code>a</code> limit, when the distribution approaches a normal
distribution.</dd>


<dt>labels</dt>
<dd>optional character vector with coefficient labels. If specified,
it should have the same length as at least one of <code>mean</code> and <code>precision</code>,
and in that case the MLiG prior with these parameters is assigned to these coefficients,
while any coefficients not present in labels will be assigned a non-informative
prior with mean 0 and precision 0.</dd>


<dt>a</dt>
<dd>scalar parameter that controls how close the prior is to independent
normal priors with <code>mean</code> and <code>precision</code> parameters. The larger
this value (default is 1000), the closer.</dd>

</dl>

pr_MLiG: Create an object representing a Multivariate Log inverse Gamma (MLiG) prior distribution

Description

Usage

Value

Arguments

References