Feature selection step using Wilcoxon test — step_select

Select variables with the lowest (adjusted) p-value of a Wilcoxon-Mann-Whitney test against an outcome.

Usage

step_select_wilcoxon(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  outcome = NULL,
  n_kept = NULL,
  prop_kept = NULL,
  cutoff = NULL,
  correction = "none",
  res = NULL,
  skip = FALSE,
  id = rand_id("select_wilcoxon")
)

# S3 method for step_select_wilcoxon
tidy(x, ...)

Arguments

recipe: A recipe object. The step will be added to the sequence of operations for this recipe.
...: One or more selector functions to choose variables for this step. See selections() for more details.
role: Not used by this step since no new variables are created.
trained: A logical to indicate if the quantities for preprocessing have been estimated.
outcome: Name of the variable to perform the test against.
n_kept: Number of variables to keep.
prop_kept: A numeric value between 0 and 1 representing the proportion of variables to keep. n_kept and prop_kept are mutually exclusive.
cutoff: Threshold beyond which (below or above) the variables are discarded.
correction: Multiple testing correction method. One of p.adjust.methods. Default to "none".
res: This parameter is only produced after the recipe has been trained.
skip: A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.
id: A character string that is unique to this step to identify it.
x: A step_select_wilcoxon object.

Value

An updated version of recipe with the new step added to the sequence of any existing operations.

Author

Antoine Bichat

Examples

rec <-
  iris %>%
  dplyr::filter(Species != "virginica") %>%
  recipe(formula = Species ~ .) %>%
  step_select_wilcoxon(all_numeric_predictors(), outcome = "Species",
                       correction = "fdr", prop_kept = 0.5) %>%
  prep()
rec
#> 
#> ── Recipe ──────────────────────────────────────────────────────────────────────
#> 
#> ── Inputs 
#> Number of variables by role
#> outcome:   1
#> predictor: 4
#> 
#> ── Training information 
#> Training data contained 100 data points and no incomplete rows.
#> 
#> ── Operations 
#> • Wilcoxon filtering against Species on: Sepal.Length, ... | Trained
tidy(rec, 1)
#> # A tibble: 4 × 5
#>   terms              pv       qv kept  id                   
#>   <chr>           <dbl>    <dbl> <lgl> <chr>                
#> 1 Sepal.Length 8.35e-14 1.11e-13 FALSE select_wilcoxon_HIyvQ
#> 2 Sepal.Width  2.14e-13 2.14e-13 FALSE select_wilcoxon_HIyvQ
#> 3 Petal.Length 5.65e-18 1.13e-17 TRUE  select_wilcoxon_HIyvQ
#> 4 Petal.Width  2.28e-18 9.14e-18 TRUE  select_wilcoxon_HIyvQ
bake(rec, new_data = NULL)
#> # A tibble: 100 × 3
#>    Petal.Length Petal.Width Species
#>           <dbl>       <dbl> <fct>  
#>  1          1.4         0.2 setosa 
#>  2          1.4         0.2 setosa 
#>  3          1.3         0.2 setosa 
#>  4          1.5         0.2 setosa 
#>  5          1.4         0.2 setosa 
#>  6          1.7         0.4 setosa 
#>  7          1.4         0.3 setosa 
#>  8          1.5         0.2 setosa 
#>  9          1.4         0.2 setosa 
#> 10          1.5         0.1 setosa 
#> # ℹ 90 more rows