NYC Crash Data

code.R

library(tidyverse)
library(patchwork)

crashes <- read_csv("~/Downloads/Motor_Vehicle_Collisions_-_Crashes.csv")
vehicles <- read_csv("~/Downloads/Motor_Vehicle_Collisions_-_Vehicles.csv")

vehicles %>% 
  group_by(COLLISION_ID) %>% 
  summarise(
    out_of_state = sum(STATE_REGISTRATION != "NY", na.rm = TRUE),
    unknown = max(is.na(STATE_REGISTRATION), na.rm = TRUE)
  ) %>% 
  ungroup() ->
  state_info

crashes %>% 
  mutate(
    is_fatal = `NUMBER OF PERSONS KILLED` > 0,
    has_injury = `NUMBER OF PERSONS INJURED` > 0
  ) %>% View()
  select(COLLISION_ID, is_fatal, `CRASH DATE`, has_injury) %>% 
  inner_join(state_info) %>% 
  mutate(has_out_of_state = out_of_state > 0) ->
  crash_data

crash_data %>% 
  na.omit() %>% 
  mutate(
    date = as.Date(`CRASH DATE`, format = "%m/%d/%Y"),
    week = lubridate::floor_date(date, "month"),
    has_out_of_state = ifelse(has_out_of_state, "Accident had Out of State Driver", "No out of state driver")
  ) %>% 
  group_by(has_out_of_state, week) %>% 
  summarise(k = mean(is_fatal)) %>% 
  ggplot(aes(week, k, color = has_out_of_state)) +
  geom_point() +
  geom_smooth(method = "gam", formula = y ~ s(x, bs = "gp", k = 50)) +
  theme_minimal() +
  facet_wrap(~ has_out_of_state, ncol = 2) +
  theme(legend.position = "none") +
  scale_y_continuous(labels = scales::percent) +
  labs(
    x = "Month",
    y = "% Crashes with Fatalities"
  ) -> g1
  
crash_data %>% 
  na.omit() %>% 
  mutate(
    date = as.Date(`CRASH DATE`, format = "%m/%d/%Y"),
    week = lubridate::floor_date(date, "month"),
    has_out_of_state = ifelse(has_out_of_state, "Accident had Out of State Driver", "No out of state driver")
  ) %>% 
  group_by(has_out_of_state, week) %>% 
  summarise(k = mean(has_injury)) %>% 
  ggplot(aes(week, k, color = has_out_of_state)) +
  geom_point() +
  geom_smooth(method = "gam", formula = y ~ s(x, bs = "gp", k = 50)) +
  theme_minimal() +
  facet_wrap(~ has_out_of_state, ncol = 2) +
  theme(legend.position = "none") +
  scale_y_continuous(labels = scales::percent) +
  labs(
    x = "Month",
    y = "% Crashes with Injuries"
  ) -> g2

g1 / g2