Estimate specific variable importance parameters

Functions that set default values to measure variable importance via specific parameters.

vimp_accuracy()

Nonparametric Intrinsic Variable Importance Estimates: Classification accuracy

vimp_anova()

Nonparametric Intrinsic Variable Importance Estimates: ANOVA

vimp_auc()

Nonparametric Intrinsic Variable Importance Estimates: AUC

vimp_deviance()

Nonparametric Intrinsic Variable Importance Estimates: Deviance

vimp_rsquared()

Nonparametric Intrinsic Variable Importance Estimates: R-squared

Estimate variable importance

General functions to measure population variable importance and compute point estimates and confidence intervals.

vim()

Nonparametric Intrinsic Variable Importance Estimates and Inference

cv_vim()

Nonparametric Intrinsic Variable Importance Estimates and Inference using Cross-fitting

vimp_se()

Estimate variable importance standard errors

vimp_ci()

Confidence intervals for variable importance

vimp_hypothesis_test()

Perform a hypothesis test against the null hypothesis of \(\delta\) importance

sp_vim()

Shapley Population Variable Importance Measure (SPVIM) Estimates and Inference

spvim_ics()

Influence function estimates for SPVIMs

spvim_se()

Standard error estimate for SPVIM values

sample_subsets()

Create necessary objects for SPVIMs

measure_anova()

Estimate ANOVA decomposition-based variable importance.

bootstrap_se()

Compute bootstrap-based standard error estimates for variable importance

Predictiveness measures

Functions to measure population predictiveness (e.g., R-squared or AUC) and compute point estimates and confidence intervals for predictiveness.

measure_accuracy()

Estimate the classification accuracy

measure_auc()

Estimate area under the receiver operating characteristic curve (AUC)

measure_cross_entropy()

Estimate the cross-entropy

measure_deviance()

Estimate the deviance

measure_mse()

Estimate mean squared error

measure_r_squared()

Estimate R-squared

est_predictiveness_cv()

Estimate a nonparametric predictiveness functional using cross-fitting

est_predictiveness()

Estimate a nonparametric predictiveness functional

Aggregate variable importance estimates

Functions to aggregate individual variable importance objects (created with calls to vim or cv_vim) by averaging or merging into larger objects.

average_vim()

Average multiple independent importance estimates

merge_vim()

Merge multiple vim objects into one

Internal package utility functions

Functions for internal convenience, used by other core functions.

print(<vim>)

Print a vim object

format(<vim>)

Format a vim object

make_folds()

Create Folds for Cross-Fitting

create_z()

Create complete-case outcome, weights, and Z

make_kfold()

Turn folds from 2K-fold cross-fitting into individual K-fold folds

run_sl()

Run a Super Learner for the provided subset of features

check_fitted_values()

Check pre-computed fitted values for call to vim, cv_vim, or sp_vim

check_inputs()

Check inputs to a call to vim, cv_vim, or sp_vim

extract_sampled_split_predictions()

Extract sampled-split predictions from a CV.SuperLearner object

get_cv_sl_folds()

Get a numeric vector with cross-validation fold IDs from CV.SuperLearner

get_full_type()

Obtain the type of VIM to estimate using partial matching

scale_est()

Return an estimator on a different scale

Deprecated functions

Functions that are kept for backwards compatability; we recommend using the updated or replacement functions instead.

vimp_regression()

Nonparametric Intrinsic Variable Importance Estimates: ANOVA

Package documentation homepage

Landing page for R help

vimp

vimp: Perform Inference on Algorithm-Agnostic Intrinsic Variable Importance

VRC01 data

Dataset for use in the vignette examples

vrc01

Neutralization sensitivity of HIV viruses to antibody VRC01