age_race_deaths.Rmd

---
title: "R Notebook"
output: html_notebook
---

```{r}
library(dplyr)
library(lubridate)
library(tidyr)
library(ggplot2)
raceage <- readr::read_csv("https://data.cdc.gov/api/views/qfhf-uhaa/rows.csv?accessType=DOWNLOAD&bom=true&format=true%20target=")
pop <- readr::read_csv("https://www2.census.gov/programs-surveys/popest/datasets/2010-2019/national/totals/nst-est2019-alldata.csv")
```

```{r}
raceage %>%
  filter(Jurisdiction != "United States") %>%
  left_join(pop, by = c("Jurisdiction" = "NAME")) %>%
  mutate(Date = mdy(`Week Ending Date`),
         DeathsPC = `Number of Deaths` / POPESTIMATE2019,
         Group = `Race/Ethnicity`) %>%
  filter(Date < max(Date) - weeks(1)) %>%
  group_by(Quarter = floor_date(Date, "quarter"), Group) %>%
  summarise(DeathsPC = sum(DeathsPC, na.rm = TRUE), .groups = "drop") %>%
  mutate(Year = year(Quarter),
         Quarter = month(Quarter)) %>%
  pivot_wider(names_from = Year, values_from = DeathsPC) %>%
  mutate(change = `2021` / `2020`) %>%
  select(-starts_with("2")) %>%
  ggplot(aes(Quarter, change, fill = Group)) +
  geom_col(position = "dodge")
```