Starting an EDA on the mult_wave part of the citizen_shield project and looking into a bunch of networks linking the determinants with the least amount of missing data

.github/workflows/sphinx.yml

@@ -0,0 +1,33 @@
+name: Sphinx build
+
+on:
+  push:
+    branches:
+      - 'main'
+
+jobs:
+  build:
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [ubuntu-latest]
+    steps:
+    - uses: actions/checkout@v2
+    - name: Build HTML
+      uses: ammaraskar/sphinx-action@0.4
+      with:
+        # docs-folder: "docs/"
+        # build-command: "sphinx-build -b html docs/source/ docs/build/html"
+        pre-build-command: "apt-get update -y && apt-get install -y pandoc"
+    - name: Upload artifacts
+      uses: actions/upload-artifact@v1
+      with:
+        name: html-docs
+        path: docs/build/html/
+    - name: Deploy
+      uses: peaceiris/actions-gh-pages@v3
+      if: github.ref == 'refs/heads/main'
+      with:
+        github_token: ${{ secrets.GITHUB_TOKEN }}
+        publish_dir: docs/build/html

.gitignore

@@ -134,3 +134,7 @@ dmypy.json
 data
 kp_determinants
 kp_determinants2
+.Rproj.user
+
+*.Rproj
+.Rhistory

citizen_shield/citizen_shield_networks.Rmd

@@ -0,0 +1,403 @@
+---
+title: "unified SEM and Multilevel VAR Estimation all determinants"
+output: html_notebook
+---
+
+# Outline
+This script covers the following steps
+
+A. N = 1 model - unified Structural Equation Model (uSEM)
+
+- Set up the uSEM model
+
+- Check model summary (model fit statistics)
+
+- Data visualization and interpretation of results
+
+B. N = all Multilevel VAR Model
+
+- Set up the mlVAR model
+
+- Check model summary(model fit statistics)
+
+- Data visualization and interpretation of results
+
+C. Selected Readings
+
+- for uSEM model: Yang, X., Ram, N., Gest, S., Lydon, D., Conroy, D. E., Pincus, A. L., & Molenaar, P. C. M. (2018). Socioemotional dynamics of emotion regulation and depressive symptoms: A person-specific network approach. Complexity, 2018, Article ID 5094179. doi: 10.1155/2018/5094179 [Open Access https://www.hindawi.com/journals/complexity/2018/5094179/]
+
+- for mlVAR model: Bringmann LF, Vissers N, Wichers M, Geschwind N, Kuppens P, Peeters F, et al. (2013) A Network Approach to Psychopathology: New Insights into Clinical Longitudinal Data. PLoS ONE 8(4): e60188. https://doi.org/10.1371/journal.pone.0060188 [Open Access https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0060188]
+
+### Preliminaries
+#### Loading Libraries
+
+Loading libraries used in this script
+
+```{r}
+# Check to see if necessary packages are installed, and install if not
+packages <- c("psych", "pompom", "mlVAR", "bootnet", "psychonetrics", "GGMncv", "EstimateGroupNetwork", "relaimpo")
+if (length(setdiff(packages, rownames(installed.packages()))) > 0) {
+  install.packages(setdiff(packages, rownames(installed.packages())))  
+}
+# Load packages
+library(tidyr)
+library(lubridate)
+library(stringr)
+library(psych) #for data description
+library(plyr) #for data manipulation
+library(dplyr)
+library(ggplot2) #for data visualization
+# library(pompom) #for uSEM
+# library(mlVAR) #for mlVAR models
+library(bootnet)
+library(psychonetrics)
+library(EstimateGroupNetwork)
+```
+
+### Loading Data
+
+```{r}
+fs_demo_columns <- c('id', 'demographic_age', 'demographic_age_factor',
+       'demographic_education', 'demographic_gender',
+       'demographic_gender_factor', 'demographic_income',
+       'demographic_living_with', 'demographic_region',
+       # 'demographic_underage_children',
+       'demographic_underage_children_factor', 'fsd_end', 
+       # 'fsd_id',
+       # 'fsd_no', 
+       'fsd_round', 
+       'fsd_start', 'fsd_vnk', 
+       # 'fsd_vr',
+       'fsd_weight')
+
+```
+
+```{r}
+df <- read.csv("../citizen_shield/data/kp_df_eng_ordinal.csv", header=TRUE, row.names = "X")
+remove_na <- function(DF, n=0) {
+  DF[, colSums(is.na(DF)) <= n]
+}
+
+df <- remove_na(df, n=4000)
+
+```
+
+```{r}
+# describeBy(df, group="user_id")
+# tmp <- df %>%
+#   group_by(user_id) %>%
+#   summarise_at(vars(feature_list), funs(sd(., na.rm=TRUE)))
+```
+
+
+
+#### create a feature list based on the column names of the data frame
+
+```{r}
+feature_list <- colnames(df %>% select(-fs_demo_columns))
+```
+
+#### Show the head and description of the overall data frame
+
+```{r}
+head(df)
+```
+
+```{r}
+# describeBy(df, group = "user_id")
+# describe(df)
+```
+
+#### Show the head and descrption of an example user 
+
+```{r}
+example_round <- 1
+data_indiv <- df[df$fsd_round == example_round, ]
+# head(data_indiv)
+# describe(data_indiv)
+```
+
+
+```{r}
+plot_df <- data_indiv %>%
+  select(c(id, fsd_round), all_of(feature_list[1:20])) %>%
+  gather(key = "variable", value = "value", -c(id, fsd_round))
+```
+
+#### plotting intraindividual change 
+
+```{r}
+#plotting intraindividual change 
+ggplot(data = plot_df,
+       aes(x = id, y=value, group= fsd_round)) +
+  #first variable
+  geom_line(aes(color = variable)) + 
+  #plot layouts
+  scale_x_continuous(name="Arbitrary Time") +
+  scale_y_continuous(name="Raw Values") +  
+  theme_classic() +
+  theme(axis.title=element_text(size=14),
+        axis.text=element_text(size=14),
+        plot.title=element_text(size=14, hjust=.5)) +
+  ggtitle(example_round)
+```
+#### Normalize the data frame (consider min max maybe)
+
+```{r}
+# standardize specific data columns (not the id or time variables in first 3 columns)
+data_indiv[feature_list] <- lapply(data_indiv[feature_list], 
+                             function(x) c(scale(x, center=TRUE, scale=TRUE)))
+# describe(data_indiv)
+```
+
+#### plotting normalized intraindividual change 
+```{r}
+plot_df <- data_indiv %>%
+  select(c(id, fsd_round), all_of(feature_list[1:20])) %>%
+  gather(key = "variable", value = "value", -c(id, fsd_round))
+```
+
+
+```{r}
+#plotting intraindividual change 
+ggplot(data = plot_df,
+       aes(x = id, y=value, group= fsd_round)) +
+  #first variable
+  geom_line(aes(color = variable)) + 
+  #plot layouts
+  scale_x_continuous(name="Arbitrary Time") +
+  scale_y_continuous(name="Raw Values") +  
+  theme_classic() +
+  theme(axis.title=element_text(size=14),
+        axis.text=element_text(size=14),
+        plot.title=element_text(size=14, hjust=.5)) +
+  ggtitle(example_round)
+```
+
+
+Now we see that all the variables are in standardized form.
+
+### Check the data
+
+It is useful to check that there are data in all columns. If any one of the variables is all missing (or has no variance), the model cannot be fit. Missing data on a few observations within a column is ok.
+
+```{r}
+# check column missing
+na_col <- 0
+for (col in 1:ncol(data_indiv)) {
+  if (sum(is.na(data_indiv[,col])) == nrow(data_indiv)){
+        na_col <- na_col + 1
+  }
+}
+na_col
+```
+
+All columns are reported.
+
+### Describing the data.
+```{r}
+describe(df)
+```
+
+### Not standardizing the data
+
+For the following networks, the data are kept in their original form.
+
+```{r}
+## Impute the feature_list missingness
+# imp.cart <- mice::mice(df[, feature_list], method="cart", printFlag = FALSE)
+# df[, feature_list] <- mice::complete(imp.cart)
+complete_df <- df[complete.cases(df[, c("id", "fsd_round", feature_list)]), c("id", "fsd_round", feature_list)]
+```
+
+### ggmModSelect and EBICglasso networks
+There is no accounting for the hierarchical nature of the data here. Unregularized Gaussian Graphical Model ("ggmModSelect"; GGM) using the glasso algorithm and stepwise model selection. Gaussian Markov random field estimation using graphical LASSO and extended Bayesian information criterion ("EBICglasso") to select optimal regularization parameter.
+
+```{r}
+net_modSelect <- estimateNetwork(complete_df[feature_list], 
+              default = "ggmModSelect",
+              stepwise = FALSE,
+              corMethod = "cor")
+```
+
+```{r}
+net_thresh <- estimateNetwork(complete_df[feature_list],
+              tuning = 0, # EBICglasso sets tuning to 0.5 by default
+              default = "EBICglasso",
+              threshold = TRUE,
+              corMethod = "cor")
+```
+
+```{r}
+Layout <- qgraph::averageLayout(net_modSelect, net_thresh)
+layout(t(1:2))
+plot(net_modSelect, layout = Layout, title = "ggmModSelect", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+plot(net_thresh, layout = Layout, title = "Thresholded EBICglasso", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+```
+
+Here the principal direction is forced - this rescales variables according to the sign of the first eigen-vector. This will lead to most correlations to be positive (positive manifold), leading to negative edges to be substantively interpretable. (not sure this is preferable in this instance as the variables are not all from the same questionnaire).
+
+```{r}
+net_modSelect_rescale <- estimateNetwork(complete_df[feature_list], 
+              default = "ggmModSelect",
+              stepwise = FALSE,
+              principalDirection = TRUE)
+net_thresh_rescale <- estimateNetwork(complete_df[feature_list],
+              tuning = 0, 
+              default = "EBICglasso",
+              threshold = TRUE,
+              principalDirection = TRUE)
+layout(t(1:2))
+plot(net_modSelect_rescale, layout = Layout, 
+     title = "ggmModSelect", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+plot(net_thresh_rescale, layout = Layout, 
+     title = "Thresholded EBICglasso", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+```
+```{r}
+qgraph::centralityPlot(
+  list(
+    ggmModSelect = net_modSelect_rescale,
+    EBICGlasso_thresh = net_thresh_rescale
+  ), include = "ExpectedInfluence"
+)
+```
+
+```{r}
+boots <- bootnet(net_thresh_rescale, statistics = "ExpectedInfluence", 
+                 nBoots = 100, nCores = 2, type = "case")
+plot(boots, statistics = "ExpectedInfluence") + 
+  theme(legend.position = "none")
+```
+### relative importance network
+There is no accounting for the hierarchical nature of the data here
+
+```{r}
+# net_relimp <- estimateNetwork(complete_df[feature_list],
+#               default = "relimp",
+#               normalize = FALSE)
+# net_relimp2 <- estimateNetwork(complete_df[feature_list],
+#               default = "relimp",
+#               normalize = FALSE,
+#               structureDefault = "ggmModSelect",
+#               stepwise = FALSE # Sent to structureDefault function
+#     )
+```
+
+
+```{r}
+# Layout <- qgraph::averageLayout(net_relimp, net_relimp2)
+# layout(t(1:2))
+# plot(net_relimp, layout = Layout, title = "Saturated", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038")
+# plot(net_relimp2, layout = Layout, title = "Non-saturated", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038")
+```
+### graphicalVAR
+This approach is not really valid as it uses all time points as if they are from the same person which is not the case here
+
+```{r}
+# Estimate model:
+gvar <- estimateNetwork(
+  complete_df, default = "graphicalVAR", vars = feature_list,
+  tuning = 0, dayvar = "fsd_round", nLambda = 10
+)
+```
+
+
+```{r}
+Layout <- qgraph::averageLayout(gvar$graph$temporal,
+                                gvar$graph$contemporaneous)
+layout(t(1:2))
+plot(gvar, graph = "temporal", layout = Layout,
+     title = "Temporal", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+plot(gvar, graph = "contemporaneous", layout = Layout,
+     title = "Contemporaneous", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+```
+
+```{r}
+# gvar_boot <- bootnet(gvar, nBoots = 2, nCores = 2)
+```
+
+```{r}
+# plot(gvar_boot, graph = "contemporaneous", plot = "interval")
+```
+
+### Mixed Graphical model
+
+```{r}
+net_mgm <- estimateNetwork(complete_df[feature_list],
+              default = "mgm",
+              type="g",
+              level=1
+              # type=c("g", "g", "g", "g", "c", "c", "c", "g"),
+              # level=c(1, 1, 1, 1, 12, 7, 15, 1)
+              )
+```
+
+```{r}
+plot(net_mgm, layout = Layout,
+     title = "mgm", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+```
+
+```{r}
+par(mar=c(5,1,10,1)+7)
+for (checkpoint in unique(complete_df[, "fsd_round"])) {
+  net_mgm <- estimateNetwork(complete_df[complete_df["fsd_round"]==checkpoint, feature_list], 
+              default = "mgm",
+              type="g",
+              level=1)
+  plot(net_mgm, layout = Layout, 
+       # title = paste("mgm network", "checkpoint =", checkpoint), 
+       edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+  
+  title(main=paste("Mixed Graphical Model network", "checkpoint =", checkpoint), line=16.25)
+}
+```
+
+
+### Partial Correlation network
+
+```{r}
+net_par_cor <- estimateNetwork(complete_df[feature_list], 
+              default = "pcor")
+```
+
+```{r}
+plot(net_par_cor, layout = Layout, 
+     title = "Partial Correlation", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+```
+
+```{r}
+par(mar=c(5,1,10,1)+7)
+for (checkpoint in unique(complete_df[, "fsd_round"])) {
+  net_par_cor <- estimateNetwork(complete_df[complete_df["fsd_round"]==checkpoint, feature_list], 
+              default = "pcor")
+  plot(net_par_cor, layout = Layout, 
+       edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+  
+  title(main=paste("Partial Correlation network", "checkpoint =", checkpoint), line=16.25)
+}
+```
+
+
+### Correlation network
+
+```{r}
+net_cor <- estimateNetwork(complete_df[feature_list], 
+              default = "cor")
+```
+
+```{r}
+plot(net_cor, layout = Layout, 
+     title = "Correlation", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+```
+
+### GGMncv network
+
+```{r}
+net_GGMncv <- estimateNetwork(complete_df[feature_list], 
+              default = "GGMncv")
+```
+
+```{r}
+plot(net_GGMncv, layout = Layout, 
+     title = "GGMncv", edge.labels=TRUE, posCol="#306fbe", negCol="#e58038", label.scale.equal=TRUE, label.cex=10)
+```

docs/Makefile

@@ -0,0 +1,20 @@
+# Minimal makefile for Sphinx documentation
+#
+
+# You can set these variables from the command line, and also
+# from the environment for the first two.
+SPHINXOPTS    ?=
+SPHINXBUILD   ?= sphinx-build
+SOURCEDIR     = source
+BUILDDIR      = build
+
+# Put it first so that "make" without argument is like "make help".
+help:
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+.PHONY: help Makefile
+
+# Catch-all target: route all unknown targets to Sphinx using the new
+# "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
+%: Makefile
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

docs/make.bat

@@ -0,0 +1,35 @@
+@ECHO OFF
+
+pushd %~dp0
+
+REM Command file for Sphinx documentation
+
+if "%SPHINXBUILD%" == "" (
+	set SPHINXBUILD=sphinx-build
+)
+set SOURCEDIR=source
+set BUILDDIR=build
+
+%SPHINXBUILD% >NUL 2>NUL
+if errorlevel 9009 (
+	echo.
+	echo.The 'sphinx-build' command was not found. Make sure you have Sphinx
+	echo.installed, then set the SPHINXBUILD environment variable to point
+	echo.to the full path of the 'sphinx-build' executable. Alternatively you
+	echo.may add the Sphinx directory to PATH.
+	echo.
+	echo.If you don't have Sphinx installed, grab it from
+	echo.https://www.sphinx-doc.org/
+	exit /b 1
+)
+
+if "%1" == "" goto help
+
+%SPHINXBUILD% -M %1 %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
+goto end
+
+:help
+%SPHINXBUILD% -M help %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
+
+:end
+popd

docs/requirements.txt

@@ -0,0 +1,6 @@
+sphinx>=4.4.0
+sphinx-bootstrap-theme>=0.8.1
+nbsphinx>=0.8.8
+nbsphinx-link>=1.3.0
+sphinx-gallery>=0.10.1
+numpydoc>=1.2

docs/source/_static/css/custom.css

@@ -0,0 +1,79 @@
+body {
+    font-family: "Raleway", sans-serif;
+    --darkgrey: grey;
+    --white: #ffffff;
+    --greenblue: #005dc7; 
+    color: var(--greenblue);
+}
+
+h1, h2, h3, h4, h5, h6 {
+    color: var(--greenblue);
+}
+
+li {
+    color: var(--greenblue);
+}
+
+li.toctree-l1 a.reference.internal code {
+    color: var(--greenblue);
+}
+
+li .reference.internal {
+    color: var(--greenblue);
+}
+
+.nav.navbar-nav li a {
+    color: var(--darkgrey);
+}
+
+.nav.navbar-nav li a:hover {
+    color: var(--darkgrey);
+}
+
+#dLabelGlobalToc {
+    color: var(--darkgrey);
+}
+
+#dLabelLocalToc {
+    color: var(--darkgrey);
+}
+
+.genindex-jumpbox a {
+    color: var(--greenblue);
+}
+
+table td li a {
+    color: var(--greenblue);
+}
+
+.footer a {
+    color: var(--greenblue);
+}
+
+.dropdown-menu>li>a:hover, .dropdown-menu>li>a:focus {
+    color: var(--greenblue);
+    text-decoration: none;
+    background-color: var(--greenyellow);
+}
+
+code {
+    color: var(--greenblue);
+}
+
+a {
+    color: var(--darkgrey);
+}
+
+a.headerlink:hover {
+    color: var(--darkgrey);
+}
+
+.navbar-default .navbar-brand {
+    color: var(--darkgrey);
+    /* background-color: transparent; */
+}
+
+.navbar-default .navbar-brand:hover, .navbar-default .navbar-brand:focus {
+    color: var(--darkgrey);
+    background-color: transparent;
+    }

docs/source/citizen_shield_demographics_classification.nblink

@@ -0,0 +1,3 @@
+{
+  "path": "../../citizen_shield/citizen_shield_demographics_classification.ipynb"
+}

docs/source/citizen_shield_demographics_clustering.nblink

@@ -0,0 +1,3 @@
+{
+  "path": "../../citizen_shield/citizen_shield_demographics_clustering.ipynb"
+}

docs/source/conf.py

@@ -0,0 +1,88 @@
+# Configuration file for the Sphinx documentation builder.
+#
+# This file only contains a selection of the most common options. For a full
+# list see the documentation:
+# https://www.sphinx-doc.org/en/master/usage/configuration.html
+
+# -- Path setup --------------------------------------------------------------
+
+# If extensions (or modules to document with autodoc) are in another directory,
+# add these directories to sys.path here. If the directory is relative to the
+# documentation root, use os.path.abspath to make it absolute, like shown here.
+#
+import os
+import sys
+
+import sphinx_bootstrap_theme
+
+sys.path.insert(0, os.path.abspath("../../"))
+
+
+# -- Project information -----------------------------------------------------
+
+project = 'Matti Jms Collabs'
+copyright = '2022, James Twose, Matti Heino'
+author = 'James Twose, Matti Heino'
+
+
+# -- General configuration ---------------------------------------------------
+
+# Add any Sphinx extension module names here, as strings. They can be
+# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
+# ones.
+extensions = ["sphinx.ext.duration",
+    "sphinx.ext.doctest",
+    "sphinx.ext.autodoc",
+    "sphinx.ext.autosummary",
+    "sphinx.ext.mathjax",
+    "sphinx.ext.viewcode",
+    "numpydoc",
+    "sphinx.ext.inheritance_diagram",
+    "nbsphinx",
+    "nbsphinx_link",
+    "sphinx_gallery.load_style",
+]
+
+# Add any paths that contain templates here, relative to this directory.
+templates_path = ['_templates']
+
+# List of patterns, relative to source directory, that match files and
+# directories to ignore when looking for source files.
+# This pattern also affects html_static_path and html_extra_path.
+exclude_patterns = ["_build", "**.ipynb_checkpoints"]
+
+
+# -- Options for HTML output -------------------------------------------------
+
+# The theme to use for HTML and HTML Help pages.  See the documentation for
+# a list of builtin themes.
+#
+html_theme = "bootstrap"
+html_theme_path = sphinx_bootstrap_theme.get_html_theme_path()
+
+html_theme_options = {
+    "source_link_position": "footer",
+    "bootswatch_theme": "simplex",
+    "navbar_title": "Matti and James Collaboration Code",
+    "navbar_sidebarrel": False,
+    "bootstrap_version": "3",
+    "nosidebar": True,
+    "body_max_width": "100%",
+    "navbar_links": [
+        # ("Gallery", "examples/index"),
+        # ("Tutorial", "tutorial"),
+        # ("Home", "index"),
+        # ("API", "api"),
+    ],
+}
+
+# Add any paths that contain custom static files (such as style sheets) here,
+# relative to this directory. They are copied after the builtin static files,
+# so a file named "default.css" will overwrite the builtin "default.css".
+html_static_path = ['_static']
+
+# Add the 'copybutton' javascript, to hide/show the prompt in code
+# examples, originally taken from scikit-learn's doc/conf.py
+def setup(app):
+    app.add_js_file("copybutton.js")
+    app.add_css_file("css/custom.css")

docs/source/index.rst

@@ -0,0 +1,50 @@
+.. Matti Jms Collabs documentation master file, created by
+   sphinx-quickstart on Tue Jun 28 20:18:11 2022.
+   You can adapt this file completely to your liking, but it should at least
+   contain the root `toctree` directive.
+
+Welcome to Matti Jms Collabs's documentation!
+=============================================
+
+.. toctree::
+   :maxdepth: 2
+   :caption: Contents:
+
+
+Citizen Shield
+----------------------------
+.. nbgallery::
+   :caption: Example Notebooks
+   :name: notebook-gallery
+   :glob:
+
+   citizen_shield_demographics_classification.ipynb
+   citizen_shield_demographics_clustering.ipynb
+
+.. raw:: html
+
+   <div id="content">
+      <li><a href="../../../citizen_shield/citizen_shield_networks.nb.html">Citizen Shield Network Approach in R</a></li>
+   </div>
+
+|
+
+References
+^^^^^^^^^^^
+.. raw:: html
+
+   <div id="content">
+      <li>Borsboom, D., Deserno, M. K., Rhemtulla, M., Epskamp, S., Fried, E. I., McNally, R. J., ... & Waldorp, L. J. (2021). 
+      Network analysis of multivariate data in psychological science.
+      <i>Nature Reviews Methods Primers, 1</i> (1), 1-18.
+      <a href="https://www.nature.com/articles/s43586-021-00055-w.pdf"> Article Link</a></li>
+   </div>
+
+|
+
+Indices and tables
+==================
+
+* :ref:`genindex`
+* :ref:`modindex`
+* :ref:`search`