In-class Exercise 2

Getting Started

Installing and Loading R Packages

Loading R packages

pacman::p_load(sf, sfdep, tmap, tidyverse, knitr)

The Data

For this In-class Exercise, the Hunan data sets will be used. They are:

1. hunan: a geographical data set in ESRI shapefile format
2. Hunan_2012: an attribute data set in csv format

Importing geospatial data

st_read() can be used to read the shape file data set into an R sf dataframe

hunan <- st_read(dsn = 'data/geospatial',
                 layer = 'Hunan')

Reading layer `Hunan' from data source 
  `D:\phlong2023\ISSS624\In-Class_Ex\In-Class_Ex2\data\geospatial' 
  using driver `ESRI Shapefile'
Simple feature collection with 88 features and 7 fields
Geometry type: POLYGON
Dimension:     XY
Bounding box:  xmin: 108.7831 ymin: 24.6342 xmax: 114.2544 ymax: 30.12812
Geodetic CRS:  WGS 84

Importing attribute table

read_csv() can be used to read the attribute file into an R data frame

hunan_2012 <- read_csv('data/aspatial/Hunan_2012.csv')

Combining the two data sets

left_join() can be used to combine the two data sets

Note

In order to retain the geospatial properties, the left data frame must be sf data.frame, in this case it is hunan

hunan_GDPPC <- left_join(hunan, hunan_2012,
                            by = 'County')%>%
  select(1:4, 7, 15) #Retaining the city's name, ID, county name, county type, GDPPC, and geometry

Deriving Contiguity Weights: Queen’s Model

The sfdep method entails the creation of a tibble data frame which contains the original data as well as the neighbors list and weights for each polygon , as opposed to creating the contiguity and weight separately in spdep.

wm_q <- hunan_GDPPC %>%
  mutate(nb = st_contiguity(geometry), # Default is Queen
         wt = st_weights(nb,
                         style='W'),
         .before = 1)

Computing Global Moran’s I (old spdep method)

moranI <- global_moran(wm_q$GDPPC,
                       wm_q$nb,
                       wm_q$wt)

moranI

$I
[1] 0.30075

$K
[1] 7.640659

Computing Local Moran’s I (with sfdep method)

lisa <- wm_q %>%
  mutate(local_moran = local_moran(GDPPC, nb, wt, nsim = 99),
         .before = 1) %>%
  unnest(local_moran)