get_varimp {sgboost}

Variable importance of a sparse-group boosting model

Description

Variable importance is computed as relative reduction of loss-function attributed to each predictor (groups and individual variables). Returns a list of two data.frames. The first contains the variable importance of a sparse-group model in a data.frame for each predictor. The second one contains the aggregated relative importance of all groups vs. individual variables.

Usage

get_varimp(sgb_model)

Arguments

sgb_model

Model of type mboost to compute the variable importance for.

Value

List of two data.frames. ⁠$raw⁠ contains the name of the variables, group structure and variable importance on both group and individual variable basis. ⁠$group_importance⁠ contains the the aggregated relative importance of all group baselearners and of all individual variables.

See Also

mboost::varimp() which this function uses.

Examples

library(mboost)
library(dplyr)
set.seed(1)
df <- data.frame(
  x1 = rnorm(100), x2 = rnorm(100), x3 = rnorm(100),
  x4 = rnorm(100), x5 = runif(100)
)
df <- df %>%
  mutate_all(function(x) {
    as.numeric(scale(x))
  })
df$y <- df$x1 + df$x4 + df$x5
group_df <- data.frame(
  group_name = c(1, 1, 1, 2, 2),
  var_name = c("x1", "x2", "x3", "x4", "x5")
)

sgb_formula <- as.formula(create_formula(alpha = 0.3, group_df = group_df))
sgb_model <- mboost(formula = sgb_formula, data = df)
sgb_varimp <- get_varimp(sgb_model)