Cramer's V of Categorical Random Forest predictions vs. observations

Fits a univariate random forest model y ~ x with the character or factor response y and the numeric, character or factor predictor x using ranger::ranger() and returns the Cramer's V (see cor_cramer()) between the observed responses and the model predictions. Cases are weighted with case_weights() to prevent issues arising from class imbalance.

Cases are weighted with case_weights() to prevent issues arising from class imbalance.

Supports cross-validation via the arguments arguments cv_training_fraction (numeric between 0 and 1) and cv_iterations (integer between 1 and n) introduced via ellipsis (...). See preference_order() for further details.

Usage

f_categorical_rf(df, ...)

Arguments

df

(required, dataframe) with columns:

• x: (numeric) numeric, character, or factor predictor.

• y (numeric) character or factor response.

...

(optional) Accepts the arguments cv_training_fraction (numeric between 0 and 1) and cv_iterations (integer between 1 and Inf) for cross validation.

Value

numeric or numeric vector: Cramer's V

See also

Other preference_order_functions: f_binomial_gam(), f_binomial_glm(), f_binomial_rf(), f_count_gam(), f_count_glm(), f_count_rf(), f_numeric_gam(), f_numeric_glm(), f_numeric_rf(), preference_order()

Examples

data(vi_smol, package = "spatialData")

df <- data.frame(
  y = vi_smol[["vi_factor"]],
  x = vi_smol[["soil_type"]]
)

#no cross-validation
f_categorical_rf(df = df)
#> [1] 0.4672802

#cross-validation
f_categorical_rf(
  df = df,
  cv_training_fraction = 0.5,
  cv_iterations = 10
  )
#>  [1] 0.4542574 0.4698654 0.4501119 0.4086847 0.4306535 0.4283688 0.4753788
#>  [8] 0.4458692 0.4788032 0.4696353

#numeric predictor
df <- data.frame(
  y = vi_smol[["vi_categorical"]],
  x = vi_smol[["swi_max"]]
)

f_categorical_rf(df = df)
#> [1] 0.531158