gs3_log

This function simulates from the conditional posterior distribution of a log transformation of the
latent U.

internal

Bayesian nonparametric density estimation modeling mixtures by a Ferguson-Klass type algorithm for posterior normalized random measures.

Guillaume Kon Kam King

BNPdensity

Ferguson-Klass Type Algorithm for Posterior Normalized Random
Measures

Julyan Arbel

Ernesto Barrios

Antonio Lijoi

Luis E. Nieto-Barajas

Igor Prünster

gs3_log function

<dl><dt>logut</dt>
<dd>Real, log of the latent variable U at the current iteration.</dd>
<dt>n</dt>
<dd>Integer, number of data points.</dd>
<dt>r</dt>
<dd>Integer, number of clusters.</dd>
<dt>alpha</dt>
<dd>Positive real. Total mass of the centering measure.</dd>
<dt>kappa</dt>
<dd>Positive real. A parameter of the NRMI process.</dd>
<dt>gama</dt>
<dd>Real. \(0\leq \texttt{gama} \leq 1\). See details.</dd>
<dt>delta</dt>
<dd>Scale of the Metropolis-Hastings proposal distribution</dd></dl>

Arguments

Conditional posterior distribution of latent logU — gs3_log

<dl>

<dt>logut</dt>
<dd>Real, log of the latent variable U at the current iteration.</dd>


<dt>n</dt>
<dd>Integer, number of data points.</dd>


<dt>r</dt>
<dd>Integer, number of clusters.</dd>


<dt>alpha</dt>
<dd>Positive real. Total mass of the centering measure.</dd>


<dt>kappa</dt>
<dd>Positive real. A parameter of the NRMI process.</dd>


<dt>gama</dt>
<dd>Real. \(0\leq \texttt{gama} \leq 1\). See details.</dd>


<dt>delta</dt>
<dd>Scale of the Metropolis-Hastings proposal distribution</dd>

</dl>

gs3_log: Conditional posterior distribution of latent logU

Description

Usage

Arguments