Parallel coordinates plot of a DEA panel over time

Draws a parallel-coordinates view of panel (multi-period) DEA data: the periods form the parallel axes, and each DMU is one line tracing the chosen quantity across time. The quantity on the vertical axis is either a single input/output variable (raw values) or the per-period DEA efficiency score.

Usage

plot_panel_io_parcoo(
  panel_data,
  y,
  inputs = NULL,
  outputs = NULL,
  id = "Label",
  period = "Period",
  orientation = "in",
  color_by_trend = FALSE,
  labels = "none",
  max.overlaps.value = 10,
  transparency = 0.7,
  fade = TRUE,
  x_angle = NULL,
  subtitle = NULL,
  title = NULL,
  interactive = FALSE,
  ...
)

Arguments

panel_data

A data frame in long format (one row per DMU-period), or a dea_panel_data object – in which case inputs, outputs, id and period are taken from the object and those arguments are ignored.

y

What to trace over time: "crs" or "vrs" for the per-period efficiency score, or the name or integer position of a single input/output column in panel_data. (If a data column is literally named "crs" or "vrs", pass its position instead.)

inputs, outputs

Optional column selectors (names or integer positions) identifying the inputs and outputs, used when y is an efficiency model and column names are not i_/o_ prefixed.

id, period

Columns (name or position) identifying the DMU and the period. Defaults "Label" and "Period".

orientation

Measurement orientation for the efficiency scores, passed to compute_efficiency (default "in").

color_by_trend

Colour each DMU's line by its overall change over the panel (last observed value minus first). FALSE (default) uses a single colour; TRUE maps the change onto the package's sequential (viridis) palette; "diverging" uses a scale anchored at zero built from the package's status colours (orange for increases, grey for no change, sky blue for decreases).

labels

Which DMUs to label at the right-hand axis: "none" (default), "all", "id", "max.overlaps", or the name of a single DMU (also highlights that DMU's line).

max.overlaps.value

Passed to ggrepel when labels = "max.overlaps" (default 10).

transparency

Opacity of the lines, a single number in [0, 1] (default 0.7).

fade

Controls the single-DMU focus view. When one DMU is given to labels, the other lines are faded so the chosen DMU stands out. TRUE (default) uses a sensible fade level; FALSE disables it; a single number in [0, 1] sets the alpha of the faded lines directly, where larger values fade them less.

x_angle

Angle in degrees for the x-axis tick labels, useful when the period labels are long. NULL (default) keeps them horizontal.

subtitle

Optional subtitle shown beneath the title. NULL (default) shows a note naming what is traced; pass a string to override, or NA to suppress.

title

Optional plot title.

interactive

Logical; static ggplot2 (default) or interactive plotly.

...

Additional arguments passed to geom_line.

Value

A ggplot2 object, or a plotly object when interactive = TRUE.

Details

When y is "crs" or "vrs", efficiency is computed separately for each period's cross-section (each period is its own DEA problem), matching plot_panel_io_biplot. When y names an input or output column, its raw values are plotted, so the axes share the variable's natural scale.

See also

plot_panel_io_biplot, plot_io_parcoo, compute_efficiency

Examples

pd <- dea_panel_data(taiwanese_banks, inputs = 3:5, outputs = 6:8,
                     id = "DMU", period = "Year")
plot_panel_io_parcoo(pd, y = "vrs")

# per-period VRS efficiency of every bank over 2009-2011
plot_panel_io_parcoo(
  taiwanese_banks, y = "vrs", id = "DMU", period = "Year",
  inputs = 3:5, outputs = 6:8
)

# colour lines by each bank's overall change in efficiency
plot_panel_io_parcoo(
  taiwanese_banks, y = "vrs", id = "DMU", period = "Year",
  inputs = 3:5, outputs = 6:8, color_by_trend = TRUE
)

# the same, on a diverging scale anchored at zero
plot_panel_io_parcoo(
  taiwanese_banks, y = "vrs", id = "DMU", period = "Year",
  inputs = 3:5, outputs = 6:8, color_by_trend = "diverging"
)

# one raw variable over time, focusing a single bank
plot_panel_io_parcoo(
  taiwanese_banks, y = "demand_deposits", id = "DMU", period = "Year",
  labels = "Cathay"
)