deploy: b561446

Author: robjhyndman

articles.html

@@ -2356,6 +2356,25 @@
 <div class="posts-container posts-apply-limit l-page">
 <div class="posts-list">
 <h1 class="posts-list-caption" data-caption="All articles">All articles</h1>
+<a href="articles/RJ-2025-006/" class="post-preview">
+<script class="post-metadata" type="text/json">{"categories":[]}</script>
+<div class="metadata">
+<div class="publishedDate">Aug. 12, 2025</div>
+<div class="dt-authors">
+<div class="dt-author">Niek Den Teuling</div>
+<div class="dt-author">Steffen Pauws</div>
+<div class="dt-author">Edwin van den Heuvel</div>
+</div>
+</div>
+<div class="thumbnail">
+<img/>
+</div>
+<div class="description">
+<h2>latrend: A Framework for Clustering Longitudinal Data</h2>
+<div class="dt-tags"></div>
+<p>Clustering of longitudinal data is used to explore common trends among subjects over time. In this paper, we focus on cases where the sole repeated measurement of interest is numeric. Various R packages have been introduced throughout the years for identifying clusters of longitudinal patterns, summarizing the variability in trajectories between subjects in terms of one or more trends. We introduce the R package latrend as a framework for the unified application of methods for longitudinal clustering, enabling comparisons between methods with minimal coding. The package also serves as an interface to commonly used packages for clustering longitudinal data, including dtwclust, flexmix, kml, lcmm, mclust, mixAK, and mixtools. This enables researchers to easily compare different approaches, implementations, and method specifications. Furthermore, researchers can build upon the standard tools provided by the framework to quickly implement new cluster methods, enabling rapid prototyping.</p>
+</div>
+</a>
 <a href="articles/RJ-2025-007/" class="post-preview">
 <script class="post-metadata" type="text/json">{"categories":[]}</script>
 <div class="metadata">
@@ -2532,25 +2551,6 @@ <h2>Space-Time Smoothing of Survey Outcomes using the R Package SUMMER</h2>
 <p>The increasing availability of complex survey data, and the continued need for estimates of demographic and health indicators at a fine spatial and temporal scale, has led to the need for spatio-temporal smoothing methods that acknowledge the manner in which the data were collected. The open source R package SUMMER implements a variety of methods for spatial or spatio-temporal smoothing of survey outcomes. In this paper, we focus primarily on demographic and health indicators. Our methods are particularly useful for data from Demographic Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS). We build upon functions within the survey package, and use INLA for fast Bayesian computation. This paper includes a brief overview of these methods and illustrates the workflow of processing surveys, fitting space-time smoothing models for both binary and composite indicators, and visualizing results with both simulated data and DHS surveys.</p>
 </div>
 </a>
-<a href="articles/RJ-2025-006/" class="post-preview">
-<script class="post-metadata" type="text/json">{"categories":[]}</script>
-<div class="metadata">
-<div class="publishedDate">Aug. 7, 2025</div>
-<div class="dt-authors">
-<div class="dt-author">Niek Den Teuling</div>
-<div class="dt-author">Steffen Pauws</div>
-<div class="dt-author">Edwin van den Heuvel</div>
-</div>
-</div>
-<div class="thumbnail">
-<img/>
-</div>
-<div class="description">
-<h2>latrend: A Framework for Clustering Longitudinal Data</h2>
-<div class="dt-tags"></div>
-<p>Clustering of longitudinal data is used to explore common trends among subjects over time. In this paper, we focus on cases where the sole repeated measurement of interest is numeric. Various R packages have been introduced throughout the years for identifying clusters of longitudinal patterns, summarizing the variability in trajectories between subjects in terms of one or more trends. We introduce the R package latrend as a framework for the unified application of methods for longitudinal clustering, enabling comparisons between methods with minimal coding. The package also serves as an interface to commonly used packages for clustering longitudinal data, including dtwclust, flexmix, kml, lcmm, mclust, mixAK, and mixtools. This enables researchers to easily compare different approaches, implementations, and method specifications. Furthermore, researchers can build upon the standard tools provided by the framework to quickly implement new cluster methods, enabling rapid prototyping.</p>
-</div>
-</a>
 <a href="articles/RJ-2025-008/" class="post-preview">
 <script class="post-metadata" type="text/json">{"categories":[]}</script>
 <div class="metadata">

articles.xml

@@ -11,7 +11,17 @@
       <link>https://journal.r-project.org/</link>
     </image>
     <generator>Distill</generator>
-    <lastBuildDate>Mon, 11 Aug 2025 00:00:00 +0000</lastBuildDate>
+    <lastBuildDate>Tue, 12 Aug 2025 00:00:00 +0000</lastBuildDate>
+    <item>
+      <title>latrend: A Framework for Clustering Longitudinal Data</title>
+      <dc:creator>Niek Den Teuling</dc:creator>
+      <dc:creator>Steffen Pauws</dc:creator>
+      <dc:creator>Edwin van den Heuvel</dc:creator>
+      <link>https://journal.r-project.org/articles/RJ-2025-006</link>
+      <description>Clustering of longitudinal data is used to explore common trends among subjects over time. In this paper, we focus on cases where the sole repeated measurement of interest is numeric. Various R packages have been introduced throughout the years for identifying clusters of longitudinal patterns, summarizing the variability in trajectories between subjects in terms of one or more trends. We introduce the R package latrend as a framework for the unified application of methods for longitudinal clustering, enabling comparisons between methods with minimal coding. The package also serves as an interface to commonly used packages for clustering longitudinal data, including dtwclust, flexmix, kml, lcmm, mclust, mixAK, and mixtools. This enables researchers to easily compare different approaches, implementations, and method specifications. Furthermore, researchers can build upon the standard tools provided by the framework to quickly implement new cluster methods, enabling rapid prototyping.</description>
+      <guid>https://journal.r-project.org/articles/RJ-2025-006</guid>
+      <pubDate>Tue, 12 Aug 2025 00:00:00 +0000</pubDate>
+    </item>
     <item>
       <title>Structured Bayesian Regression Tree Models for Estimating Distributed Lag Effects: The R Package dlmtree</title>
       <dc:creator>Seongwon Im</dc:creator>
@@ -135,16 +145,6 @@ meningitis cases in an Italian region.</description>
       <guid>https://journal.r-project.org/articles/RJ-2025-005</guid>
       <pubDate>Thu, 07 Aug 2025 00:00:00 +0000</pubDate>
     </item>
-    <item>
-      <title>latrend: A Framework for Clustering Longitudinal Data</title>
-      <dc:creator>Niek Den Teuling</dc:creator>
-      <dc:creator>Steffen Pauws</dc:creator>
-      <dc:creator>Edwin van den Heuvel</dc:creator>
-      <link>https://journal.r-project.org/articles/RJ-2025-006</link>
-      <description>Clustering of longitudinal data is used to explore common trends among subjects over time. In this paper, we focus on cases where the sole repeated measurement of interest is numeric. Various R packages have been introduced throughout the years for identifying clusters of longitudinal patterns, summarizing the variability in trajectories between subjects in terms of one or more trends. We introduce the R package latrend as a framework for the unified application of methods for longitudinal clustering, enabling comparisons between methods with minimal coding. The package also serves as an interface to commonly used packages for clustering longitudinal data, including dtwclust, flexmix, kml, lcmm, mclust, mixAK, and mixtools. This enables researchers to easily compare different approaches, implementations, and method specifications. Furthermore, researchers can build upon the standard tools provided by the framework to quickly implement new cluster methods, enabling rapid prototyping.</description>
-      <guid>https://journal.r-project.org/articles/RJ-2025-006</guid>
-      <pubDate>Thu, 07 Aug 2025 00:00:00 +0000</pubDate>
-    </item>
     <item>
       <title>CDsampling: An R Package for Constrained D-Optimal Sampling in Paid Research Studies</title>
       <dc:creator>Yifei Huang</dc:creator>

articles/RJ-2025-002/RJ-2025-002.pdf

@@ -450,7 +450,7 @@ \subsection{Fitting methods}\label{fitting-methods}
 #> 
 #> Cluster sizes (K=2):
 #>           A           B 
-#> 135 (44.9%) 166 (55.1%) 
+#> 166 (55.1%) 135 (44.9%) 
 #> 
 #> Number of obs: 3913, strata (Patient): 301
 #> 
@@ -578,8 +578,8 @@ \subsubsection{Assessing a cluster result}\label{assessing-a-cluster-result}
 
 \begin{verbatim}
 #>        Fit
-#> 1 2.919423
-#> 2 2.024865
+#> 1 4.249461
+#> 2 5.815110
 \end{verbatim}
 
 The \texttt{predictPostprob()} and \texttt{predictAssignments()} functions compute the posterior probability and cluster membership for new trajectories, respectively. As this is not a common use case for cluster methods, most of the underlying packages do not provide this functionality. For demonstration purposes, we have implemented the functionality for the \texttt{lcModelKML} class.
@@ -626,12 +626,12 @@ \subsubsection{Identifying the number of clusters}\label{identifying-the-number-
 
 \begin{verbatim}
 #>         Dunn      WMAE estimationTime
-#> 1         NA 1.4261264          0.450
-#> 2 0.10737225 0.7850566          0.668
-#> 3 0.09944419 0.6523208          0.892
-#> 4 0.11353357 0.6081128          1.142
-#> 5 0.13487175 0.5598639          1.424
-#> 6 0.13196444 0.5209264          1.599
+#> 1         NA 1.4261264          0.472
+#> 2 0.10737225 0.7850566          0.696
+#> 3 0.09944419 0.6523208          0.968
+#> 4 0.11353357 0.6081128          1.176
+#> 5 0.13487175 0.5598639          1.311
+#> 6 0.13196444 0.5209264          1.746
 \end{verbatim}
 
 As the preferred solution corresponds to the highest Dunn index, we can obtain the respective model by calling the \texttt{max()} function on the \texttt{lcModels} list object.