descriptive.Rmd

---
title: "Descriptive analysis of cholera outbreaks in SSA"
output: html_notebook
editor_options: 
  chunk_output_type: console
---

## Outbreak characteristics

```{r}
library(data.table)
d = fread("https://raw.githubusercontent.com/kimfinale/cholera_outbreaks_ssa/main/reference_data/outbreak_data.csv")

unique(d$pop_cat)
cat = c("Population <10,000", "Population <100,000", "Population <1,000,000", "Population >1,000,000")

ar = lapply(cat, function(x) d[d$pop_cat==x,]$attack_rate)
len = sapply(ar, length)
df = data.frame(attack_rate=do.call('c',ar), pop_size=rep(cat,len))

#
summary(df$attack_rate)
hist(df$attack_rate)
library(ggplot2)
library(tidyverse)
df %>% 
  ggplot()+
  geom_histogram(aes(attack_rate,y=..density..))

df %>% 
  ggplot()+
  geom_boxplot(aes(attack_rate))+
  scale_x_continuous(trans='log10')

summary(df$attack_rate)

quantile(df$attack_rate, probs=c(0.025,0.25,0.5,0.75,0.975))


library(ggplot2)
ggplot(df)+
  geom_histogram(aes(x=attack_rate,y=..density..))+
  scale_x_binned(limits=c(0,50))+
  facet_wrap(~pop_size)+
  theme_bw()
  
dens = density(ar[[1]])
df2 = data.frame(x=dens$x, y=dens$y, pop_size=cat[1])
for (i in 2:4) {
  dens = density(ar[[i]])
  df2 = rbind(df2, data.frame(x=dens$x, y=dens$y, pop_size=cat[i]))
}

ggplot(df2, aes(x=x, ymin=0, ymax=y, fill=pop_size))+
  geom_ribbon(alpha=0.5)+
  scale_x_continuous(limits=c(0,50))+
  facet_wrap(~pop_size)+
  theme_bw()


```