New R -package This Usupot: Visualization of the decomposition of differences in speed meter | R-Bloggers

3.1 Prepare data for

To make theseus plots, you need two data frames that share common columns.

We use the New York City flight data from 2013 from NYCFLIGHTS13 As a demo set set. Here we will define the speed matric if the share of flights that arrived on time. In December 2013, the per-time arrival percentage fell considerably compared to November. We are investigating the cause using a Thoseus -Plot.

First we create one on_time Column in the data frame to indicate whether every flight has arrived on time. We then extract the flights for November and December in separate data frames to form two comparison groups. The arrival percentage on De Tijd was 64% in November and fell to 47% in December.

library(dplyr)
library(nycflights13)

data <- flights |> 
  filter(!is.na(arr_delay)) |>
  mutate(on_time = arr_delay <= 0) |>  # Arrived on time
  left_join(airlines, by = "carrier") |>
  mutate(carrier = name) |>  # Convert carrier abbreviations to full names
  select(year, month, day, origin, dest, carrier, dep_delay, on_time)

data |> head()
#> # A tibble: 6 × 8
#>    year month   day origin dest  carrier                dep_delay on_time
#>                                  
#> 1  2013     1     1 EWR    IAH   United Air Lines Inc.          2 FALSE  
#> 2  2013     1     1 LGA    IAH   United Air Lines Inc.          4 FALSE  
#> 3  2013     1     1 JFK    MIA   American Airlines Inc.         2 FALSE  
#> 4  2013     1     1 JFK    BQN   JetBlue Airways               -1 TRUE   
#> 5  2013     1     1 LGA    ATL   Delta Air Lines Inc.          -6 TRUE   
#> 6  2013     1     1 EWR    ORD   United Air Lines Inc.         -4 FALSE

data_Nov <- data |> filter(month == 11)
data_Dec <- data |> filter(month == 12)

data_Nov |> summarise(on_time_rate = mean(on_time)) |> pull(on_time_rate)
#> [1] 0.6426161
data_Dec |> summarise(on_time_rate = mean(on_time)) |> pull(on_time_rate)
#> [1] 0.4672835

3.2 Basics

With the help of the two prepared data frames, we first make one ship object. The ship Object is a copy of the R6 class ShipOfTheseusDesigned to suddenly make theseus.

library(TheseusPlot)

ship <- create_ship(data_Nov, data_Dec, y = on_time, labels = c("November", "December"))

You can make a theseus plot by passing on column names to the plot method of one ship object. For example, to make a theseus plot for the airport of origin:

ship$plot(origin)

New York City has three major airports and Newark Liberty International Airport (EWR) was good for the majority of the fall in the temporary arrival percentage.

Note that the number of flights at every airport cases, because a larger flight volume is expected to have a greater impact. To make this clear, the Theseus -Plot shows the data size for each group within each subgroup as a bar chart. We see from this that the number of flights is comparable to the airports, which makes a direct comparison of contributions possible.

In summary, a theseus -plot consists of two components:

• A waterfall plot shows how much each subgroup has contributed to the change in the metric.
• A bar chart that represents the sample size for each group within each subgroup.

A ship Object also offers the table Method to inspect the exact values ​​that are used in the Thisus Plot.

ship$table(origin)
#> # A tibble: 3 × 8
#>   origin contrib    n1    n2    x1    x2 rate1 rate2
#>             
#> 1 EWR    -0.0831  9603  9410  6251  3901 0.651 0.415
#> 2 JFK    -0.0565  8645  8923  5702  4332 0.660 0.485
#> 3 LGA    -0.0358  8723  8687  5379  4393 0.617 0.506

3.3 Turn the plot around

When there are many subgroups, a theseus plot can be difficult to read. In such cases you can exchange the X and Y axes for better visualization.

ship$plot_flip(carrier)

When the number of subgroups is large, those with small contributions are automatically grouped. This happens as standard when there are more than 10 subgroups, but the threshold can be adjusted with the n argument.

ship$plot_flip(carrier, n = 5)

From this plot, Jetblue Airways and United Air Lines seem to have the biggest contributions to the decrease in the temporary arrival percentage.

3.4 Automatic discretization of continuous values

Theseus suddenly do not immediately support continuous variables. If a continuous column is provided, it is automatically discreted. For example, we can make a theseus -plot for delays delays.

ship$plot_flip(dep_delay)

Continuous variables are discreted as standard so that each subgroup has approximately the same sample sizes, with the number of bins set at 10. You can change these settings by giving the return value of continuous_config() at the continuous argument.

ship$plot_flip(dep_delay, continuous = continuous_config(n = 5))

This result shows that both a decrease in the departure of time and an increase in delayed departure has contributed to the fall in the temporary arrival percentage.