cluster_map: Create Cluster Map Based on Local Moran's I

In DHSr: Create Large Scale Repeated Regression Summary Statistics Dataset and Visualization Seamlessly

Create Cluster Map Based on Local Moran's I

Description

This function creates a map of clusters based on Local Moran's I values. It identifies clusters using Queen contiguity and visualizes them on a map.

Usage

cluster_map(
  dataset,
  lisa_value,
  lisa_label,
  label,
  lisa_cutoff,
  location_var,
  location_name,
  level2 = NULL,
  id_start = 0,
  comparison = ">",
  min_area = 0,
  min_ = 5,
  title = "Clusters Based on Queen Contiguity",
  subtitle = "",
  footnote = "",
  legend_position = "bottom",
  color_scheme = "C"
)

Arguments

dataset	A spatial dataset of class 'sf'.
lisa_value	The name of the variable in the dataset containing Local Moran's I values.
lisa_label	The name of the variable in the dataset containing the LISA label.
label	The specific label to filter clusters.
lisa_cutoff	A numeric value specifying the cutoff for LISA values.
location_var	The variable name indicating the primary location in the dataset.
location_name	The name of the variable for the location names.
level2	An optional second level of location hierarchy. Default is 'NULL'.
id_start	The starting value for cluster IDs. Default is '0'.
comparison	The comparison operator for filtering ('>', '<', '>=', etc.). Default is '>'.
min_area	Minimum area required for a cluster to be considered valid. Default is '0'.
min_	Minimum number of districts required for a cluster to be valid. Default is '5'.
title	The title of the map. Default is '"Clusters Based on Queen Contiguity"'.
subtitle	A subtitle for the map. Default is '""'.
footnote	A footnote for the map. Default is '""'.
legend_position	The position of the legend on the map. Default is '"bottom"'.
color_scheme	The color scheme for the map. Default is '"C"'.

Value

A list with the following components:

dataset_with_clusters	An 'sf' object containing the dataset with assigned cluster IDs.
summary_clusters	A data frame summarizing cluster information, including regions, area, and the number of locations.
plot	A 'ggplot' object for visualizing the clusters.

Examples

if (requireNamespace("spData", quietly = TRUE)) {
  library(sf)
  library(dplyr)
  library(spdep)
  library(spData)
  library(ggplot2)

  # Load US states data from spData
  us_states <- spData::us_states

  # Simplify for demonstration: Select a subset of columns
  us_states_data <- us_states %>%
    select(GEOID, NAME) %>%
    mutate(mean_wealth = rnorm(nrow(us_states), 50, 10))  # Add mock data

  # Define a shapefile path
  shapefile_path <- tempfile(fileext = ".shp")
  sf::st_write(us_states, shapefile_path, quiet = TRUE)

  # Corrected listw function call using your package
  us_states_listw <- DHSr::listw(
    shapefile_path = shapefile_path,
    data = us_states_data %>% sf::st_drop_geometry(),  # Drop geometry for compatibility
    loc_shape = "GEOID",
    loc_data = "GEOID",
    weight_function = function(d) exp(-d / 0.2)
  )

  # Apply Spdeplisa function
  lisa_result <- DHSr::Spdeplisa(
    data = us_states_data,
    variable_name = "mean_wealth",
    listw = us_states_listw
  )

  # Add LISA labels
  lisa_result <- lisa_result %>%
    mutate(lisa_label = case_when(
      lisa_I > 0 ~ "High-High",
      lisa_I < 0 ~ "Low-Low",
      TRUE ~ "Others"
    ))

  # Apply cluster_map function
  cluster_map_result <- DHSr::cluster_map(
    dataset = lisa_result,
    lisa_value = "lisa_I",
    lisa_label = "lisa_label",
    label = "High-High",
    lisa_cutoff = 0.5,
    location_var = "GEOID",
    location_name = "NAME",
    id_start = 1,
    comparison = ">",
    min_area = 0,
    min_ = 3,  # Reduced for smaller demonstration
    title = "Clusters Based on Queen Contiguity",
    subtitle = "High-High Clusters",
    footnote = "Generated using DHSr package",
    legend_position = "bottom",
    color_scheme = "C"
  )

  # View the resulting dataset with clusters
  head(cluster_map_result$dataset_with_clusters)

  # View the cluster summary
  print(cluster_map_result$summary_clusters)

  # Plot the clusters
  print(cluster_map_result$plot)
}

DHSr documentation built on April 4, 2025, 12:18 a.m.