Our World in Data data transformation for this notebook https://observablehq.com/@nd/recreating-covid-19-cases-chart

covid_data_deaths_cases.R

library(readr)
library(dplyr)
library(tidyr)
library(caTools)
options(scipen = 999999)

df <- read_csv("~/Downloads/owid-covid-data.csv")

countries_to_use <- c(
  "Austria",
  "Czech Republic",
  "France",
  "Germany",
  "Greece",
  "Hungary",
  "Netherlands",
  "Poland",
  "Russia",
  "Slovakia",
  "Slovenia",
  "Sweden"
)

df <- df %>%
  filter(location %in% countries_to_use) %>%
  select(location, date, total_cases, new_cases, total_deaths, new_deaths) %>%
  group_by(location) %>%
  mutate(
    days_since_100 = as.numeric(date - min(date[total_cases >= 100])),
    days_since_5 = as.numeric(date - min(date[total_deaths >= 5]))
  ) %>%
  ungroup()

cases <- df %>%
  select(location, date, total_cases, new_cases, days_since_100) %>%
  group_by(location) %>%
  filter(days_since_100 >= 0) %>%
  mutate(
    rolling3 = runmean(new_cases, 3),
    rolling7 = runmean(new_cases, 7),
    rolling14 = runmean(new_cases, 14),
    rolling30 = runmean(new_cases, 30)
  ) %>%
  ungroup()
  

deaths <- df %>%
  select(location, date, total_deaths, new_deaths, days_since_5) %>%
  group_by(location) %>%
  filter(days_since_5 >= 0) %>%
  mutate(
    rolling3 = runmean(new_deaths, 3),
    rolling7 = runmean(new_deaths, 7),
    rolling14 = runmean(new_deaths, 14),
    rolling30 = runmean(new_deaths, 30)
  ) %>%
  ungroup()