scrapes covidtracking.com and plots tests, cases, and deaths over time

usa-covid-19.R

library(dplyr)
library(ggplot2)
library(tidyr)
library(stringr)
library(rvest)

usa_html <- read_html("https://covidtracking.com/data/us-daily")

usa_dat <- usa_html %>% html_table() %>% `[[`(1)

usa_dat <- 
  usa_dat %>% 
  rename_all(tolower) %>%
  mutate(
    date = as.Date(date, "%a %B %d %Y")
  ) %>%
  mutate_at(
    vars(-c(date)), .funs = function(x) {as.numeric(str_replace_all(x, "[[:punct:]]", ""))}
  ) %>% 
  arrange(date) %>% 
  mutate(
    new_positive = c(NA, diff(positive)),
    new_deaths = c(NA, diff(deaths))
  )

usa_dat %>% ggplot(aes(x = date, y = new_deaths)) + geom_point() + geom_line()
usa_dat %>% ggplot(aes(x = date, y = `new tests`)) + geom_point() + geom_line()
usa_dat %>% ggplot(aes(x = date, y = new_positive / `new tests`)) + geom_point() + geom_line()


usa_dat %>% 
  dplyr::select(date, new_deaths, new_positive, `new tests`) %>% 
  pivot_longer(cols = c(new_deaths, new_positive, `new tests`), names_to = "metric") %>%
  group_by(metric) %>%
  mutate(
    scaled_value = (value - mean(value, na.rm = T)) / sd(value, na.rm = T)
  ) %>%
  ggplot(aes(x = date, y = scaled_value, color = metric)) + geom_point(alpha = .3) + geom_line(alpha = .3) + geom_smooth(se = F)