vimp: vimp: Perform Inference on Algorithm-Agnostic Intrinsic...

In bdwilliamson/nova: Perform Inference on Algorithm-Agnostic Variable Importance

vimp: Perform Inference on Algorithm-Agnostic Intrinsic Variable Importance

Description

A unified framework for valid statistical inference on algorithm-agnostic measures of intrinsic variable importance. You provide the data, a method for estimating the conditional mean of the outcome given the covariates, choose a variable importance measure, and specify variable(s) of interest; 'vimp' takes care of the rest.

Author(s)

Maintainer: Brian Williamson https://bdwilliamson.github.io/ Contributors: Jean Feng https://www.jeanfeng.com, Charlie Wolock https://cwolock.github.io/

Methodology authors:

• Brian D. Williamson

• Jean Feng

• Peter B. Gilbert

• Noah R. Simon

• Marco Carone

See Also

Manuscripts:

• \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1111/biom.13392")}

  (R-squared-based variable importance)

• \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1111/biom.13389")}

  (Rejoinder to discussion on R-squared-based variable importance article)

• http://proceedings.mlr.press/v119/williamson20a.html (general Shapley-based variable importance)

• \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1080/01621459.2021.2003200")}

  (general variable importance)

Other useful links:

• https://bdwilliamson.github.io/vimp/

• https://github.com/bdwilliamson/vimp

• Report bugs at https://github.com/bdwilliamson/vimp/issues

Imports

The packages that we import either make the internal code nice (dplyr, magrittr, tibble, rlang, MASS, data.table), are directly relevant to estimating the conditional mean (SuperLearner) or predictiveness measures (ROCR), or are necessary for hypothesis testing (stats) or confidence intervals (boot, only for bootstrap intervals).

We suggest several other packages: xgboost, ranger, gam, glmnet, polspline, and quadprog allow a flexible library of candidate learners in the Super Learner; ggplot2 and cowplot help with plotting variable importance estimates; testthat, WeightedROC, cvAUC, and covr help with unit tests; and knitr, rmarkdown, and tidyselect help with the vignettes and examples.

Author(s)

Maintainer: Brian D. Williamson [e-mail] (ORCID)

Other contributors:

• Jean Feng [contributor]

• Charlie Wolock [contributor]

• Noah Simon (ORCID) [thesis advisor]

• Marco Carone (ORCID) [thesis advisor]

See Also

Useful links:

• https://bdwilliamson.github.io/vimp/

• https://github.com/bdwilliamson/vimp

• http://bdwilliamson.github.io/vimp/

• Report bugs at https://github.com/bdwilliamson/vimp/issues

bdwilliamson/nova documentation built on Feb. 1, 2024, 10:04 p.m.