05_milk_industry.Rmd

name: milkindustry
# Milk Industry

What's a milk cow worth? This data comes via the #TidyTuesday project.  I constructed a waterfall plot to track the average price of milk cows, calculating percent changes year on year. 

```{r, echo = F}

dir <- "raw_data"
file <- paste0(dir, "/", "milkcow_facts.csv")
url <- "https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2019/2019-01-29/milkcow_facts.csv"

# create raw_data folder
if (!dir.exists(dir)) { dir.create(dir) }

# download data from internet and save
if (!file.exists(file)) { download.file(url = url, destfile = file) }

# read in downloaded data
milkcow_facts <- readr::read_csv(file)
```



```{r, echo = F}
df <- milkcow_facts %>% 
  mutate(milk_cow_cost_per_animal_lag =
           lag(milk_cow_cost_per_animal)) %>% 
  mutate(between_cost = (milk_cow_cost_per_animal_lag + milk_cow_cost_per_animal)/2) %>% 
  mutate(percent_change = 
           round(100*(milk_cow_cost_per_animal - milk_cow_cost_per_animal_lag) / milk_cow_cost_per_animal_lag, 0)) 

df_changes <- df %>% filter(year > 2004)
```

A random sample from the data set: 
```{r, echo = F}
knitr::kable(head(df, 5), format = "html")
```


---


```{r milk, eval=F, echo=F}
  ggplot(df_changes) +
  aes(xmin = year - .3, 
      xmax = year + .3) + 
  aes(ymin = milk_cow_cost_per_animal_lag, 
      ymax = milk_cow_cost_per_animal) +
  geom_rect(fill = "blue", alpha = .3) +
  geom_col(data = df %>% filter(year == 2004), 
           mapping = aes(x = year, y = milk_cow_cost_per_animal), 
           fill = "grey", alpha = .5) +
  labs(x = "") +
  labs(fill = "") +
  labs(y = "Cost (US$)") +
  labs(title = "Cost of milk cows in the United States") +
  labs(subtitle = "Per animal cost, 2004-2014 | Viz: Gina Reynolds") +
  scale_y_continuous(limits = c(0, 2004)) +
  scale_x_continuous(breaks = seq(2004, 2014, by = 2), 
                     limits = c(2003.5, 2014.5)) +
  aes(x = year - .5,  
      xend = year + 1.5) +
  aes(y = milk_cow_cost_per_animal, 
      yend = milk_cow_cost_per_animal) +
  geom_segment(data = df %>% filter(year >= 2004 & year < 2014),
               linetype = "dashed", col = "gray39") +
  geom_segment(data = df %>% filter(year == 2014),
               linetype = "dashed", col = "gray39",
               mapping = aes(x = year - .5, xend = year + .5)) +
  scale_fill_viridis_d() +
  theme_bw(base_family = "Times") +
  aes(label = paste0(percent_change, "%")) +
  geom_text(mapping = aes(y = milk_cow_cost_per_animal, 
                          x = year, 
                          col = percent_change > 0),  
            size = 3, nudge_y = 90 * ((df_changes$percent_change > 0) - .5),
            show.legend = F) +
  scale_color_manual(values = c("tomato4", "gray39")) +
  labs(x = "") +
  labs(y = "Price ($US)")
```

```{r, echo = F, warning=F, message=F, eval = T, fig.show='hide'}
get_what_save_what <- "milk"
eval(parse(text = paste(knitr:::knit_code$get(get_what_save_what), collapse = "")))
ggsave(paste0("figures/", get_what_save_what, ".png"), dpi = 300)
```

`r apply_reveal("milk")`

---