Skip to contents

hib_calc_aoa: Calculate the Area-of-Applicability.

Source: R/hib_calc_aoa.R
hib_calc_aoa.Rd

Calculate the Area-of-Applicability (AOA) given a model, cross-validation output, and the data being used to make projection/prediction.

Usage

hib_calc_aoa(mod_obj, cv_obj, proj_data)

Arguments

mod_obj

One of; HibModelLME4, HibModelINLA or HibModelBRMS.

cv_obj

A HibCV object, based on the above mod_obj.

proj_data

A data.frame, SpatRaster or list of SpatRaster, containing covariates for projecting/predicting the model. Predicted responses should not be included.

Value

list

Details

This function estimates an AOA, for each of the cross-validation approaches supplied. We follow the methods of Meyer and Pebesma 2021, to estimate a dissimilarity index (DI) using model fitting/projection variables, and cross-validation folds to estimate applicability limits. See references for further information.
The returned list includes:

  • variable_weights, a data.frame of variable importance weights.

  • variable_scale_params, a list of the parameters used to scale each variable.

  • train_DI, a list containing DI estimates and thresholds per cross-validation type.

  • proj_aoa, the AOA output, where 1 indicates a location within the AOA, 0 outside the AOA, same data type as proj_data.

References

  • Meyer, H., & Pebesma, E. (2021). Predicting into unknown space? Estimating the area of applicability of spatial prediction models. Methods in Ecology and Evolution, 12, 1620–1633. https://doi.org/10.1111/2041-210X.13650

  • Meyer H, Milà C, Ludwig M, Linnenbrink J (2023). CAST: 'caret' Applications for Spatial-Temporal Models. R package, https://CRAN.R-project.org/package=CAST.

See also

hib_cv_model HibCV

Examples

if (FALSE) { # \dontrun{
  aoa_res <- hib_calc_aoa(mod_obj, cv_obj, proj_data)
} # }