Chapter 6 EDA Flight Delays

data(nycflights)

names(nycflights) ?nycflights

glimpse(nycflights)

ggplot(data = nycflights, aes(x = dep_delay)) + geom_histogram()

ggplot(data = nycflights, aes(x = dep_delay)) + geom_histogram(binwidth = 15)

ggplot(data = nycflights, aes(x = dep_delay)) + geom_histogram(binwidth = 150)

#exercise 1- The three histograms show the same data but look different because of the bin width. The histogram with a 15-minute binwidth shows more detail in the distribution of delays, while the 150-minute binwidth groups too many values together and hides important patterns. The default histogram falls somewhere in between, showing the overall shape without as much detail as the smaller bins.

lax_flights <- nycflights |> filter(dest == “LAX”) ggplot(data = lax_flights, aes(x = dep_delay)) + geom_histogram()

lax_flights |> summarise(mean_dd = mean(dep_delay), median_dd = median(dep_delay), n = n()) sfo_feb_flights <- nycflights |> filter(dest == “SFO”, month == 2)

#Exercise 2- 68 flights meet this criteria

#Exercise 3- The shape of the histogram peaks before 0 and declines from there.

ggplot(data = sfo_feb_flights, aes(x = arr_delay)) + geom_histogram(binwidth = 15)

sfo_feb_flights |> group_by(origin) |> summarise(median_dd = median(dep_delay), iqr_dd = IQR(dep_delay), n_flights = n())

nycflights |> group_by(month) |> summarise(mean_dd = mean(dep_delay)) |> arrange(desc(mean_dd))

sfo_feb_flights |> group_by(carrier) |> summarise(median_dd = median(arr_delay), iqr_dd = IQR(arr_delay), n_flights = n())

#Exercise 4- American Airlines had the most variable arrival delays with 5

nycflights |> group_by(month) |> summarise(mean_dd = mean(dep_delay)) |> arrange(desc(mean_dd))

#Exercise 5- Choosing the month with the lowest mean departure delay accounts for all flights, including rare extreme delays, but can be skewed by outliers. Choosing the month with the lowest median reflects the typical flight experience and ignores extreme delays, though it might miss occasional very long delays.

nycflights <- nycflights |> mutate(dep_type = ifelse(dep_delay < 5, “on time”, “delayed”))

nycflights |> group_by(origin) |> summarise(ot_dep_rate = sum(dep_type == “on time”) / n()) |> arrange(desc(ot_dep_rate))

#Exercise 6- Based on the data, LaGuardia Airport has the most on time departures.

#Exercise 7- Average speed is calculated as distance ÷ (air_time / 60) to convert minutes into hours and produce speed in mph.

nycflights <- nycflights %>% mutate(avg_speed = distance / (air_time / 60))

ggplot(data = nycflights, aes(x = distance, y = avg_speed)) + geom_point()

#Exercise 8- The scatterplot shows that average speed generally increases as distance increases

nycflights %>% filter(carrier %in% c(“AA”, “DL”, “UA”)) %>% ggplot(aes(x = dep_delay, y = arr_delay, color = carrier)) + geom_point()

#Exercise 9- From the scatterplot, it appears that flights with a departure delay of about 20–30 minutes or less can still sometimes arrive on time because flights may make up time in the air.