国会选区作为 R 中的地图？

问题描述

目前，我正在使用 United States Congressional District Shapefiles。我想使用 cartogram package 中列出的步骤制作 cartogram。但我似乎无法成功地使用 cartogram_cont 函数将其放入制图对象。

您能提供的任何帮助、建议或见解都将非常有帮助。

请在代码底部指出错误发生的位置。

谢谢！

######################### Library #####################
library(sf)
library(tmap)
library(cartogram)
library(ggplot2)
library(ggmap)
library(maptools)


###################### Get Congress Map ###################
get_congress_map <- function(cong=113) {
  tmp_file <- tempfile()
  tmp_dir  <- tempdir()
  zp <- sprintf("http://cdmaps.polisci.ucla.edu/shp/districts%03i.zip",cong)
  download.file(zp,tmp_file)
  unzip(zipfile = tmp_file,exdir = tmp_dir)
  fpath <- paste(tmp_dir,sprintf("districtShapes/districts%03i.shp",cong),sep = "/")
  st_read(fpath)
}


cd114 <- get_congress_map(114)

################ Create Congressional district ID ###################

State_Dictonary <- rbind(data.frame(state.abb,STATENAME= state.name),data.frame(state.abb= "DC",STATENAME= "district Of Columbia"))

#Add DC
cd114.1 <- merge(State_Dictonary,cd114,by = c("STATENAME"))
unique(cd114$STATENAME)
#**************** Paste Zero infront of Single districts ****************
cd114.1$district_code <- as.character(cd114.1$disTRICT)
cd114.1$district_code[cd114.1$district_code == 98] <- "01" #Account for the weird code for DC
cd114.1$district_code[cd114.1$district_code == 0] <- "01"  #Account for at large districts
cd114.1$district_code[cd114.1$district_code == 1] <- "01"
cd114.1$district_code[cd114.1$district_code == 2] <- "02"
cd114.1$district_code[cd114.1$district_code == 3] <- "03"
cd114.1$district_code[cd114.1$district_code == 4] <- "04"
cd114.1$district_code[cd114.1$district_code == 5] <- "05"
cd114.1$district_code[cd114.1$district_code == 6] <- "06"
cd114.1$district_code[cd114.1$district_code == 7] <- "07"
cd114.1$district_code[cd114.1$district_code == 8] <- "08"
cd114.1$district_code[cd114.1$district_code == 9] <- "09"

#************** Create CD in  Data ****************
cd114.1$CD <- paste(cd114.1$state.abb,cd114.1$district_code,sep = "")


################ Create Fake Data TO Simulate My Own ########################
CD <- unique(cd114.1$CD)
Values <- sample(1:100,436,replace = T)

Df <- data.frame(CD,Values)


################### Merge Shape File to Donor #####################
#******************** Merge Descriptive Stats *********************
Df.2 <- merge(x = Df,y = cd114.1,by = c("CD"),all.y = T)

################### Make SF Object ####################
SF.DF.1 <- st_as_sf(Df.2)
SF.DF.2 <- st_make_valid(SF.DF.1)
class(SF.DF.2)

################### Make cartogram ####################
#Follow Proceedure Laid out: https://cran.r-project.org/web/packages/cartogram/readme/README.html

# I use st_transform instead of spTransform beause I'm using an SF Ojbect
SF.DF.3 <- st_transform(SF.DF.1,CRS("+init=epsg:3395")) 

cartogram_cont(SF.DF.3,"Value",itermax = 5)
#Error in Fij[distance <= radius[j]] <- Fbij[distance <= radius[j]] : 
#NAs are not allowed in subscripted assignments

解决方法

第 87 行中的几何条目为空 (MULTIPOLYGON EMPTY)。您可以通过以下方式解决此问题：

library(dplyr)

SF.DF.3 <- st_transform(SF.DF.1,CRS("+init=epsg:3395")) %>% 
  filter(!st_is_empty(.))

此外，您的 df 中没有列 Value，它是 Values ;)

sf_ncont <- cartogram_ncont(SF.DF.3,"Values")

tm_shape(sf_ncont) + tm_polygons("Values",style = "jenks",legend.show = FALSE) +
  tm_layout(frame = FALSE)

注意，我在下面的情节中过滤掉了阿拉斯加和夏威夷，并使用了 cartogram_ncont 而不是 cartogram_cont，因为前一个操作非常耗时。