sampleK_alpha

This sampling step only relies on the current
partition and is independent of the current component-specific
parameters, see Frühwirth-Schnatter et al (2021).

internal

Fits Bayesian finite mixtures with an unknown number of components using the telescoping sampler and different component distributions. For more details see Frühwirth-Schnatter et al. (2021) <doi:10.1214/21-BA1294>.

Gertraud Malsiner-Walli

telescope

Bayesian Mixtures with an Unknown Number of Components

Bettina Grün

Sylvia Frühwirth-Schnatter

sampleK_alpha function

<dl><dt>Kp_j</dt>
<dd>A number; indicating the current value of \(K_+\).</dd>
<dt>Kmax</dt>
<dd>A number; indicating the maximum value of \(K\) for which the conditional posterior is evaluated.</dd>
<dt>Nk_j</dt>
<dd>A numeric vector; indicating the group sizes in the partition, i.e.,
the current number of observations in the filled components.</dd>
<dt>alpha</dt>
<dd>A number; indicating the value of the parameter \(\alpha\).</dd>
<dt>log_pK</dt>
<dd>A function; evaluating the log prior of \(K\).</dd></dl>

Arguments

Sample K conditional on \(\alpha\) where \(e0 = \alpha/K\). — sampleK_alpha

<dl>

<dt>Kp_j</dt>
<dd>A number; indicating the current value of \(K_+\).</dd>


<dt>Kmax</dt>
<dd>A number; indicating the maximum value of \(K\) for which the conditional posterior is evaluated.</dd>


<dt>Nk_j</dt>
<dd>A numeric vector; indicating the group sizes in the partition, i.e.,
the current number of observations in the filled components.</dd>


<dt>alpha</dt>
<dd>A number; indicating the value of the parameter \(\alpha\).</dd>


<dt>log_pK</dt>
<dd>A function; evaluating the log prior of \(K\).</dd>

</dl>

sampleK_alpha: Sample K conditional on \(\alpha\) where \(e0 = \alpha/K\).

Description

Usage

Value

Arguments