NYC AirBnB

Does rating vary by neighborhood, room type, or both?

nyc_airbnb %>%
 group_by(neighbourhood) %>%
 summarize(median = median(stars, na.rm = TRUE), 
           mean = mean(stars, na.rm = TRUE),
           number_of_listings = n()) %>% 
  knitr::kable()

nyc_airbnb %>%
  group_by(room_type) %>%
  summarize(median_stars = median(stars, na.rm = TRUE), 
            average_stars = mean(stars, na.rm = TRUE),
           number_of_listings = n())

## # A tibble: 3 x 4
##   room_type       median_stars average_stars number_of_listings
##   <chr>                  <dbl>         <dbl>              <int>
## 1 Entire home/apt            5          4.78              26050
## 2 Private room               5          4.71              22822
## 3 Shared room                5          4.68               1169

ggplot(nyc_airbnb, aes(x = stars), color = boro) +
 geom_histogram() +
 facet_grid(~room_type)

nyc_airbnb %>%
  group_by(borough, room_type) %>%
  summarise(
  mean_stars = mean(stars, na.rm = TRUE)) %>% 
  spread(key = room_type, value = mean_stars)

## # A tibble: 5 x 4
## # Groups:   borough [5]
##   borough       `Entire home/apt` `Private room` `Shared room`
##   <chr>                     <dbl>          <dbl>         <dbl>
## 1 Bronx                      4.59           4.50          4.35
## 2 Brooklyn                   4.73           4.67          4.61
## 3 Manhattan                  4.84           4.79          4.80
## 4 Queens                     4.71           4.68          4.60
## 5 Staten Island              4.69           4.62          5

Where are rentals located?

nyc_airbnb %>% 
  filter(borough == "Manhattan", price < 1000) %>% 
  ggplot(aes(x = longitude, y = latitude, color = price)) + 
  geom_point(alpha = .2) + 
  facet_grid(~room_type) + 
  coord_cartesian() + 
  scale_color_viridis()

Entire homes are more expensive than other room types, and are more clustered in southern Manhattan.

Expensive areas

nyc_airbnb %>%
 select(borough, neighbourhood, price) %>%
 filter(borough == "Manhattan") %>%
 group_by(neighbourhood) %>%
 summarize(mean_price = mean(price, na.rm = TRUE),
           median_price = median(price, na.rm = TRUE)) %>% 
  top_n(5, median_price) %>%
 arrange(desc(median_price))

## # A tibble: 5 x 3
##   neighbourhood     mean_price median_price
##   <chr>                  <dbl>        <dbl>
## 1 Tribeca                 413.          251
## 2 NoHo                    265.          245
## 3 Battery Park City       295.          225
## 4 Flatiron District       318.          218
## 5 Midtown                 278.          210

nyc_airbnb %>%
 select(borough, neighbourhood, price) %>%
 filter(borough == "Brooklyn") %>%
 group_by(neighbourhood) %>%
 summarize(mean_price = mean(price, na.rm = TRUE),
           median_price = median(price, na.rm = TRUE)) %>% 
  top_n(5, median_price) %>%
 arrange(desc(median_price))

## # A tibble: 6 x 3
##   neighbourhood     mean_price median_price
##   <chr>                  <dbl>        <dbl>
## 1 DUMBO                   190.         162.
## 2 Vinegar Hill            164.         159 
## 3 Boerum Hill             180.         150 
## 4 Brooklyn Heights        195.         150 
## 5 Carroll Gardens         181.         150 
## 6 Downtown Brooklyn       162.         150

nyc_airbnb %>%
 select(borough, neighbourhood, price) %>%
 filter(borough == "Manhattan") %>%
 group_by(neighbourhood) %>%
 summarize(mean_price = mean(price, na.rm = TRUE),
           median_price = median(price, na.rm = TRUE)) %>% 
  top_n(-5, median_price) %>%
 arrange(median_price)

## # A tibble: 6 x 3
##   neighbourhood       mean_price median_price
##   <chr>                    <dbl>        <dbl>
## 1 Washington Heights        89.4           70
## 2 Inwood                    88.3           75
## 3 Roosevelt Island         103.            79
## 4 Marble Hill               82.2           89
## 5 Harlem                   120.            90
## 6 Morningside Heights      109.            90

nyc_airbnb %>%
 select(borough, neighbourhood, price) %>%
 filter(borough == "Brooklyn") %>%
 group_by(neighbourhood) %>%
 summarize(mean_price = mean(price, na.rm = TRUE),
           median_price = median(price, na.rm = TRUE)) %>% 
  top_n(-5, median_price) %>%
 arrange(median_price)

## # A tibble: 6 x 3
##   neighbourhood   mean_price median_price
##   <chr>                <dbl>        <dbl>
## 1 Gerritsen Beach       74           50  
## 2 Borough Park          69.7         55  
## 3 Gravesend             81.3         61  
## 4 Cypress Hills         80.2         61.5
## 5 Brownsville           86.3         65  
## 6 Bushwick              81.8         65