#install.packages("nycflights23")
#install.packages(c("viridis", "RColorBrewer", "treemap", "alluvial", "ggalluvial", "babynames"))
#install.packages("streamgraph")
library(nycflights23)
library(tidyverse)
data(flights)
data(airlines)
data(airports)Heatmaps, Treemaps, Alluvials, and Streamgraphs
Practicing visualizations
Clean the data
flights_nona <- flights |>
filter(!is.na(distance) & !is.na(arr_delay) & !is.na(dep_delay))
# remove na's for distance, arr_delay, departure delay
# add the airline names
flights2 <- left_join(flights_nona, airlines, by = "carrier")
flights2$name <- gsub("Inc\\.|Co\\.", "", flights2$name)# remove "Inc." or "Co." from the names
flights2$name <- trimws(flights2$name)# remove the extra spaces
# find the 6 airlines with the most flights
top6 <- flights2 |>
count(name) |>
arrange(desc(n)) |>
head(6)
top6# A tibble: 6 × 2
name n
<chr> <int>
1 Republic Airline 85431
2 United Air Lines 77438
3 JetBlue Airways 64280
4 Delta Air Lines 60364
5 Endeavor Air 52204
6 American Airlines 39750Summary table for each destination
by_dest <- flights_nona |>
group_by(dest) |> # group all destinations
summarise(count = n(), # counts totals for each destination
avg_dist = mean(distance), # calculates the mean distance traveled
avg_arr_delay = mean(arr_delay), # calculates the mean arrival delay
avg_dep_delay = mean(dep_delay), # calculates the mean dep delay
.groups = "drop") |> # remove the grouping structure after summarizing
arrange(avg_arr_delay) |>
filter(avg_dist < 3000)
head(by_dest)# A tibble: 6 × 5
dest count avg_dist avg_arr_delay avg_dep_delay
<chr> <int> <dbl> <dbl> <dbl>
1 PNS 71 1030 -10.6 -1.24
2 HHH 461 695. -9.95 1.38
3 HDN 27 1728 -9.93 8.78
4 VPS 107 988 -9.41 2.62
5 AVP 140 93 -8.53 -0.957
6 GSO 2857 456. -7.77 3.81 Scatterplot
ggplot(by_dest, aes(avg_dist, avg_arr_delay)) +
geom_point(aes(size = count), alpha = .3) +
geom_smooth(se = FALSE) + # remove the error band
scale_size_area() +
theme_bw() +
labs(x = "Average Flight Distance (miles)",
y = "Average Arrival Delay (minutes)",
size = "Number of Flights \n Per Destination",
caption = "Source: nycflights23 package",
title = "Average Distance and Average Arrival Delays from Flights from NY")`geom_smooth()` using method = 'loess' and formula = 'y ~ x'
Heatmaps
Heatmap 1: busiest destinations
I only kept the 25 busiest destinations so the row names are easy to read.
by_dest_top <- by_dest |>
arrange(desc(count)) |>
head(25)
by_dest_matrix <- data.matrix(by_dest_top[, -1]) # drop dest from matrix so it won't show in heatmap
row.names(by_dest_matrix) <- by_dest_top$dest # restore row names
library(viridis)Loading required package: viridisLiteby_dest_heatmap <- heatmap(by_dest_matrix,
Rowv = NA,
Colv = NA,
col = viridis(250),
cexCol = .7, # shrink x-axis label size
scale = "column",
xlab = "",
ylab = "",
main = "Heatmap of the 25 Busiest Destinations from NY")
Heatmap 2: airline and month
airline_month <- flights2 |>
filter(name %in% top6$name) |>
group_by(name, month) |>
summarise(avg_arr_delay = mean(arr_delay), .groups = "drop")
ggplot(airline_month, aes(x = month, y = name, fill = avg_arr_delay)) +
geom_tile(color = "white") +
scale_fill_viridis_c() +
scale_x_continuous(breaks = 1:12) +
theme_bw() +
labs(x = "Month (1 = January)",
y = "Airline",
fill = "Avg Arrival \n Delay (min)",
caption = "Source: nycflights23 package",
title = "Average Arrival Delay by Airline and Month")
Which 6 destination airports have the highest average arrival delay from NYC?
worst6 <- by_dest |>
arrange(desc(avg_arr_delay)) |>
head(6)
# change the airport codes to the airport names
worst6 <- left_join(worst6, airports, by = c("dest" = "faa"))
worst6 |>
select(dest, name, count, avg_arr_delay)# A tibble: 6 × 4
dest name count avg_arr_delay
<chr> <chr> <int> <dbl>
1 PSE Mercedita Airport 319 37.6
2 RNO Reno Tahoe International Airport 129 34.4
3 ABQ Albuquerque International Sunport 218 26.7
4 ONT Ontario International Airport 353 26.1
5 BQN Rafael Hernández Airport 957 25.6
6 SJU Luis Muñoz Marín International Airport 5312 21.0Treemaps
Treemap 1: carriers
The index is the carrier, the size of the box is the average distance, and the color is the average arrival delay.
flights3 <- flights2 |>
group_by(name) |>
summarise(avg_dist = mean(distance), # calculates the mean distance traveled
avg_arr_delay = mean(arr_delay)) # calculates the mean arrival delay
library(RColorBrewer)
library(treemap)
treemap(flights3,
index = "name",
vSize = "avg_dist",
vColor = "avg_arr_delay",
type = "manual",
palette = "RdYlBu", # use RColorBrewer palette
title = "Average Distance and Arrival Delay by Carrier", # plot title
title.legend = "Avg Arrival Delay (min)") # legend label
Treemap 2: airport and airline
The big rectangles are the NYC airports (EWR, JFK, LGA) and the small ones inside are the airlines. The size is the number of flights.
flights4 <- flights2 |>
filter(name %in% top6$name) |>
group_by(origin, name) |>
summarise(count = n(), .groups = "drop")
treemap(flights4,
index = c("origin", "name"), # big group first, small group second
vSize = "count",
title = "Number of Flights by Airport and Airline")
Alluvials
Alluvials need three variables: a time variable, a value, and a category.
Alluvial 1: refugees
library(alluvial)
library(ggalluvial)
data(Refugees)
ggalluv <- Refugees |>
ggplot(aes(x = year, y = refugees, alluvium = country)) +
theme_bw() +
geom_alluvium(aes(fill = country),
color = "white",
width = .1,
alpha = .8,
decreasing = FALSE) +
scale_fill_brewer(palette = "Spectral") +
# Spectral has enough colors for all countries listed
scale_x_continuous(lim = c(2002, 2013)) +
labs(title = "UNHCR-Recognised Refugees Top 10 Countries\n (2003-2013)",
y = "Number of Refugees",
fill = "Country",
caption = "Source: United Nations High Commissioner for Refugees (UNHCR)")
options(scipen = 999) # no scientific notation on the y-axis
ggalluv
Alluvial 2: flights by month
flights_month <- flights2 |>
filter(name %in% top6$name) |>
group_by(month, name) |>
summarise(flights = n(), .groups = "drop") |>
complete(month, name, fill = list(flights = 0)) # puts a 0 if an airline has no flights in a month
ggalluv_flights <- flights_month |>
ggplot(aes(x = month, y = flights, alluvium = name)) +
theme_bw() +
geom_alluvium(aes(fill = name),
color = "white",
width = .1,
alpha = .8,
decreasing = FALSE) +
scale_fill_brewer(palette = "Set2") +
scale_x_continuous(breaks = 1:12) +
labs(title = "Monthly Flights from NYC by Airline (2023)",
x = "Month (1 = January)",
y = "Number of Flights",
fill = "Airline",
caption = "Source: nycflights23 package")
ggalluv_flights