Skip to contents

Plot leave-one-out influence

Source: R/leave-out.R
plot.sdmTMB_leave_out.Rd

Plots the results of sdmTMB_leave_out(): the full index together with the index from each leave-out refit (type = "index"), the ratio of each leave-out index to the full index (type = "ratio"), or influence against haul share (type = "leverage"), where identifiers above the leverage_limit line and the influence_limit line are labelled. Years without data after removing an identifier are gaps.

Usage

# S3 method for class 'sdmTMB_leave_out'
plot(
  x,
  type = "index",
  ids = NULL,
  index_divisor = NULL,
  unit = NA,
  base_size = 11,
  influence_limit = 0.1,
  leverage_limit = 2,
  ...
)

Arguments

x

An object of class sdmTMB_leave_out.

type

Character. "index" (default), "ratio" or "leverage".

ids

Optional subset of identifiers to plot. Defaults to all; for many identifiers (e.g. hauls), pass the most influential ones.

index_divisor

Optional numeric divisor applied to index values.

unit

Character. Unit label for the index axis. Default NA.

base_size

Numeric. Base font size. Default 11.

influence_limit, leverage_limit

Numeric. Flag limits for type = "leverage". Defaults 0.1 and 2.

...

Currently ignored.

Value

A ggplot object.

See also

sdmTMB_leave_out()

Author

Mikko Vihtakari

Examples

# \donttest{
data(sebastes)
sub <- subset(sebastes, cruiseseries %in% c("EcoS", "WinterS") & year >= 2020)
mesh <- sdmTMB::make_mesh(sub, c("X", "Y"), cutoff = 60)
fit <- sdmTMB::sdmTMB(
  biomass ~ 1, data = sub, time = "year", mesh = mesh,
  family = sdmTMB::tweedie(), spatiotemporal = "iid",
  offset = log(sub$swept_area)
)
region <- make_survey_prediction_region(sub, crs = 32633)
grid <- sdmTMB::replicate_df(
  make_prediction_grid(region, resolution = 30), "year", unique(sub$year)
)
lo <- sdmTMB_leave_out(fit, id_col = "cruiseseries", newdata = grid)
#> ⠙ Refitting the full model
#> ✔ Refitting the full model [4ms]
#> 
#> ⠙ Refitting without each of 2 identifiers
#> ✔ Refitting without each of 2 identifiers [3ms]
#> 
plot(lo)

plot(lo, type = "ratio")

plot(lo, type = "leverage")

# }