Skip to contents

Plot an sdmTMB selection table

Source: R/model-selection.R
plot.sdmTMB_selection.Rd

type = "selection" draws the ranking as a status-coded dot plot: one point per candidate, positioned by how far it is from the best, coloured and shaped by whether it crashed, failed to converge or failed sdmTMB::sanity(). type = "index" computes and overlays the survey index each candidate produces, which answers the question a selection table cannot: does this choice actually move the index?

Usage

# S3 method for class 'sdmTMB_selection'
plot(
  x,
  type = "selection",
  metric = NULL,
  newdata = NULL,
  index_divisor = NULL,
  unit = NA,
  bias_correct = FALSE,
  level = 0.95,
  area = 1,
  nsplit = NULL,
  silent = TRUE,
  base_size = 11,
  ...
)

Arguments

x

An object of class sdmTMB_selection.

type

Character. "selection" (default) or "index".

metric

Metric for type = "selection": "cv_deltaNLL", "deltaAIC" or "deltaNLL". NULL (default) picks the cross-validated metric when it is available and deltaAIC otherwise.

newdata

Prediction grid, required for type = "index".

index_divisor

Optional numeric divisor applied to index values.

unit

Character. Unit label for the index axis.

bias_correct, level, area, nsplit, silent

Passed to sdmTMB::get_index_split() for type = "index".

base_size

Numeric. Base font size. Default 11.

...

Currently ignored.

Value

A ggplot object.

See also

sdmTMB_selection()

Author

Mikko Vihtakari

Examples

# \donttest{
cands <- list(
  "Year only" = density ~ 0 + as.factor(year),
  "+ depth" = density ~ 0 + as.factor(year) + depth_scaled
)
sel <- sdmTMB_selection(
  cands, data = sdmTMB::pcod, cutoff = 25, family = sdmTMB::tweedie(),
  time = "year", spatial = "on", spatiotemporal = "iid"
)
#> Candidate 'Year only': fitting.
#> Candidate '+ depth': fitting.
plot(sel)

nd <- sdmTMB::replicate_df(sdmTMB::qcs_grid, "year", unique(sdmTMB::pcod$year))
plot(sel, type = "index", newdata = nd)

# }