Add comments and clean up calculation of Dfr3.

src/Covid19MortalityWCAfrica.Rmd

@@ -147,8 +147,8 @@ Lock$name <- reorder(Lock$name, Lock$T80)
 
 
 ```{r country-data}
-## Variables relative to European countries: 
-## group, name, lockdown date, population, cumulated deaths,
+## Variables relative to African countries: 
+## group, name, lockdown date, population, cumulated deaths (as of today),
 ## and death rate.
 country_data <- 
   map_dfr(
@@ -189,6 +189,7 @@ country_data <-
 ## with population and lockdown date added
 Dfr <- deaths_by_ctry_dt %>% 
   inner_join(
+    ## filter African countries by joining
     country_data,
     by = "name"
   ) %>% 
@@ -196,8 +197,9 @@ Dfr <- deaths_by_ctry_dt %>%
     # Order name levels in decreasing order of death rate
     name = fct_reorder(name, death_rate, .desc = TRUE)
   ) %>% 
-  select(name, date, deaths, pop, lockdown)
+  select(name, date, deaths, pop, lockdown, current_cum_deaths = cum_deaths)
 
+## onam is equal to african_countries but ordered by decreasing death-rate
 onam <- levels(Dfr$name)
 
 ## Names of countries with less than 25 cumulated deaths
@@ -207,22 +209,20 @@ nam25 <- country_data %>%
 
 onam2 <- onam[!(onam %in% nam25)]
 
-## the condition on deaths >= 0 is needed because Spain has declared a
-## negative count in early June
-Dfr2 <- subset(Dfr, name %in% onam2 & deaths >= 0)
-Dfr2$name <- factor(as.character(Dfr2$name),
-                    levels = onam2)
 
-Dfr3  <- do.call(
-    "rbind",
-    by(Dfr2,
-       list(name = Dfr2$name),
-       function(x){
-           x$day = as.numeric(x$date - min(x$date, na.rm = T))
-           x
-       }))
-Dfr3$name <- factor(as.character(Dfr3$name),
-                    levels = onam2)
+## Compute day since first reported death by country, for countries with at
+## least 25 cumulative deaths
+Dfr3 <- Dfr %>% 
+  ## the condition on deaths >= 0 is needed because Spain has declared a
+  ## negative count in early June
+  filter(current_cum_deaths >= 25, deaths >= 0) %>% 
+  droplevels() %>% 
+  group_by(name) %>% 
+  mutate(
+    day = as.integer(date - min(date, na.rm = TRUE))
+  ) %>% 
+  ungroup() %>% 
+  select(-current_cum_deaths)
 
 ## use function "by" to loop over the country names
 L2 <- by(Dfr3,