# usethis::edit_r_environ() # and set STATS19_DOWNLOAD_DIRECTORY to keep the data files
vehicles = get_stats19(type = "vehicle", year = 1979)
range(vehicles$collision_year)
# [1] 1979 2024
sum(is.na(vehicles$generic_make_model)) # 0 NA values

# Number of distinct makes and models over the years
n_per_year = vehicles |>
  filter(collision_year >= 1989) |>
  group_by(collision_year) |>
  summarise(
    n_make_model = n_distinct(generic_make_model),
    percent_na = round(sum(grepl("missing", generic_make_model)) / n() * 100),
    top_5_make_model = paste(
      names(sort(table(generic_make_model), decreasing = TRUE)[2:6]),
      collapse = ", "
    )
  ) 
knitr::kable(n_per_year)