Creating a county level map of county growth rate/doubling time using R & tmap

covid_gifmap

library(tidyverse)
library(ggrepel)
library(gghighlight)
library(tidycensus)
library(sf)

dailydata<-read_csv("https://github.com/nytimes/covid-19-data/raw/master/us-counties.csv")

#This next step requires a census API 
cty_pop<-get_acs(geography="county",variables = "B01001_001",geometry=TRUE) %>%
  rename(fips=GEOID)

dailydata_ga<-dailydata %>%
  filter(state=="Georgia" &
           date>as.Date("2020-03-16")&
           county!="Unknown") 

dailydata_ga_lag<-dailydata_ga %>%
  group_by(county) %>%
  arrange(desc(date)) %>%
  mutate(value_3d=lag(x=cases,n=3,default=0,order_by=date),
         change_3d=cases-value_3d,
         double_time=if_else(value_3d==0,0,(70/(change_3d/value_3d*100)*3))) %>%
  mutate(date_c=as.character(date))

cty_pop_ga<-cty_pop %>%
  filter(substr(fips,1,2)=="13")

double_date<-function(date_sel){
  cty_select<-dailydata_ga_lag %>%
    filter(date_c==date_sel & cases>9)
  
  map<-tm_shape(cty_pop_ga)+
    tm_polygons(col="#e0e0e0",border.col = "#95989c")+
    tm_shape(cty_select)+
    tm_polygons("double_time",
                breaks=c(0,1,2,4,8,16,50),
                palette=c('#016c59','#1c9099','#67a9cf','#a6bddb','#d0d1e6','#f6eff7'),
                border.col = "#95989c",
                title="Doubling time in days \n(darker=faster growth)")+
    tm_layout(legend.outside=TRUE,
              attr.outside=TRUE,
              main.title = paste("COVID cases-growth rate in Georgia: ",date_sel,sep=""),
              main.title.size=1,
              title.position =c("left","bottom"))+
    tm_credits(text="Counties with > 10 cases shown, Data: NY Times, Map by @jerry_shannon")
  #map
  tmap_save(map,paste("ga_maps/",date_sel,".png",sep=""))
}

dates<-unique(dailydata_ga_lag$date_c)
maps<-map(dates,double_date)

mapfiles<-paste("ga_maps/",list.files(path="ga_maps"),sep="")
gifski(mapfiles,"ga_doubling_2020_04_07.gif",width=800,height=500)