updateU

Update time-constant latent node positions using optimized matrix operations.
Pre-computes invariant quantities outside the node loop for better performance.

Network changepoint analysis for undirected network data. The package implements a hidden Markov network change point model (Park and Sohn (2020)). Functions for break number detection using the approximate marginal likelihood and WAIC are also provided. This version includes performance optimizations with vectorized MCMC operations and modern ggplot2-based visualizations with colorblind-friendly palettes.

Jong Hee Park

NetworkChange

Bayesian Package for Network Changepoint Analysis

Yunkyu Sohn

updateU function

<dl><dt>K</dt>
<dd>The dimensionality of Z</dd>
<dt>U</dt>
<dd>The most recent draw of latent node positions</dd>
<dt>V</dt>
<dd>Layer-specific network generation rule</dd>
<dt>R</dt>
<dd>The dimensionality of latent space</dd>
<dt>Zb</dt>
<dd>Z - beta</dd>
<dt>s2</dt>
<dd>error variance</dd>
<dt>eU</dt>
<dd>The mean of U</dd>
<dt>iVU</dt>
<dd>The variance of U</dd></dl>

Arguments

Update time-constant latent node positions — updateU

<dl>

<dt>K</dt>
<dd>The dimensionality of Z</dd>


<dt>U</dt>
<dd>The most recent draw of latent node positions</dd>


<dt>V</dt>
<dd>Layer-specific network generation rule</dd>


<dt>R</dt>
<dd>The dimensionality of latent space</dd>


<dt>Zb</dt>
<dd>Z - beta</dd>


<dt>s2</dt>
<dd>error variance</dd>


<dt>eU</dt>
<dd>The mean of U</dd>


<dt>iVU</dt>
<dd>The variance of U</dd>

</dl>

updateU: Update time-constant latent node positions

Description

Usage

Value

Arguments