The function performs a data-driven variable selection. Starting from the provided model it iterates through all the variables starting from the one with the highest contribution (permutation importance or maxent percent contribution). If the variable is correlated with other variables (according to the given method and threshold) it performs a Jackknife test and among the correlated variables it removes the one that results in the best performing model when removed (according to the given metric for the training dataset). The process is repeated until the remaining variables are not highly correlated anymore.

varSel(
  model,
  metric,
  bg4cor,
  test = NULL,
  env = NULL,
  method = "spearman",
  cor_th = 0.7,
  permut = 10,
  use_pc = FALSE
)

Arguments

model

SDMmodel or SDMmodelCV object.

metric

character. The metric used to evaluate the models, possible values are: "auc", "tss" and "aicc".

bg4cor

SWD object. Background locations used to test the correlation between environmental variables.

test

SWD. Test dataset used to evaluate the model, not used with aicc and SDMmodelCV objects, default is NULL.

env

stack containing the environmental variables, used only with "aicc", default is NULL.

method

character. The method used to compute the correlation matrix, default "spearman".

cor_th

numeric. The correlation threshold used to select highly correlated variables, default is 0.7.

permut

integer. Number of permutations, default is 10.

use_pc

logical, use percent contribution. If TRUE and the model is trained using the Maxent method, the algorithm uses the percent contribution computed by Maxent software to score the variable importance, default is FALSE.

Value

The SDMmodel or SDMmodelCV object trained using the selected variables.

Details

  • To find highly correlated variables the following formula is used: $$| coeff | \le cor_th$$

Author

Sergio Vignali

Examples

# \donttest{ # Acquire environmental variables files <- list.files(path = file.path(system.file(package = "dismo"), "ex"), pattern = "grd", full.names = TRUE) predictors <- raster::stack(files) # Prepare presence and background locations p_coords <- virtualSp$presence bg_coords <- virtualSp$background # Create SWD object data <- prepareSWD(species = "Virtual species", p = p_coords, a = bg_coords, env = predictors, categorical = "biome")
#> Extracting predictor information for presence locations...
#> Extracting predictor information for absence/background locations...
# Split presence locations in training (80%) and testing (20%) datasets datasets <- trainValTest(data, test = 0.2, only_presence = TRUE) train <- datasets[[1]] test <- datasets[[2]] # Train a model model <- train(method = "Maxnet", data = train, fc = "l") # Prepare background locations to test autocorrelation, this usually gives a # warning message given that less than 10000 points can be randomly sampled bg_coords <- dismo::randomPoints(predictors, 10000)
#> Warning: generated random points = 0.9775 times requested number
bg <- prepareSWD(species = "Virtual species", a = bg_coords, env = predictors, categorical = "biome")
#> Extracting predictor information for absence/background locations...
#> Info: 9 absence/background locations are NA for some environmental variables, they are discarded!
if (FALSE) { # Remove variables with correlation higher than 0.7 accounting for the AUC, # in the following example the variable importance is computed as permutation # importance vs <- varSel(model, metric = "auc", bg4cor = bg, test = test, cor_th = 0.7, permut = 1) vs # Remove variables with correlation higher than 0.7 accounting for the TSS, # in the following example the variable importance is the MaxEnt percent # contribution # Train a model # The next line checks if Maxent is correctly configured but you don't need # to run it in your script if (dismo::maxent(silent = TRUE)) { model <- train(method = "Maxent", data = train, fc = "l") vs <- varSel(model, metric = "tss", bg4cor = bg, test = test, cor_th = 0.7, use_pc = TRUE) vs # Remove variables with correlation higher than 0.7 accounting for the aicc, # in the following example the variable importance is the MaxEnt percent # contribution vs <- varSel(model, metric = "aicc", bg4cor = bg, cor_th = 0.7, use_pc = TRUE, env = predictors) vs } } # }