In-class Exercise 5: Local Colocation Quotient Analysis

Published

February 6, 2023

Modified

March 6, 2023

1 Getting Started

Using Taiwanese convenience stores as a use case.

2 Installing and loading R packages

sfdep package allows us to perform colocation methods

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

studyArea <- st_read(dsn="data",
                     layer="study_area") %>% st_transform(crs=3829)

Reading layer `study_area' from data source 
  `C:\annatrw\IS415\In-class_Ex\In-class_Ex05\data' using driver `ESRI Shapefile'
Simple feature collection with 7 features and 7 fields
Geometry type: POLYGON
Dimension:     XY
Bounding box:  xmin: 121.4836 ymin: 25.00776 xmax: 121.592 ymax: 25.09288
Geodetic CRS:  TWD97

stores <- st_read(dsn="data",
                     layer="stores") %>% st_transform(crs=3829)

Reading layer `stores' from data source 
  `C:\annatrw\IS415\In-class_Ex\In-class_Ex05\data' using driver `ESRI Shapefile'
Simple feature collection with 1409 features and 4 fields
Geometry type: POINT
Dimension:     XY
Bounding box:  xmin: 121.4902 ymin: 25.01257 xmax: 121.5874 ymax: 25.08557
Geodetic CRS:  TWD97

[reference this for take home 1] [always plot polygon then points so points will not be blocked by polygon] [plotting functional and non-functional water point]

tmap_mode("view")
tm_shape(studyArea) +
  tm_polygons() +
tm_shape(stores) +
  tm_dots(col = "Name", 
          size = 0.01,
          border.col="black",
          border.lwd=0.5  )

tm_view(set.zoom.limits = c(12,16))

$tm_layout
$tm_layout$set.zoom.limits
[1] 12 16

$tm_layout$style
[1] NA


attr(,"class")
[1] "tm"

Searching for the 6 nearest neighbours (using adaptive kernel density method) - choosing 6 nearest since the target point is included in function as well (total 7 and will not get an even split)

nb <- include_self(st_knn(st_geometry(stores), 6))

wt <- st_kernel_weights(nb, stores, "gaussian", adaptive=TRUE)

Filtering out stores that are family marts and convert into a dataframe, before saving into a variable A

FamilyMart <- stores %>% filter(Name=="Family Mart")
A <-FamilyMart$Name

SevenEleven <- stores %>% filter(Name=="7-Eleven")
B <-SevenEleven$Name

A is target, B is neighbour category at 50 simulations outputs the p value directly

LCLQ <- local_colocation(A,B,nb,wt,49)

LCLQ_stores <- cbind(stores,LCLQ)

cannot do relational join on LCLQ and LCLQ_stores since there is no unique identifier of LCLQ
using cbind function only works if you do not sort the original input data
cbind inherits the properties of the first parameter

tmap_mode("view")
tm_shape(studyArea) +
  tm_polygons() +
tm_shape(LCLQ_stores) +
  tm_dots(col = "X7.Eleven",
          size = 0.01,
          border.col="black",
          border.lwd=0.5  )

tm_view(set.zoom.limits = c(12,16))

$tm_layout
$tm_layout$set.zoom.limits
[1] 12 16

$tm_layout$style
[1] NA


attr(,"class")
[1] "tm"