Projection Predictive Feature Selection
Extract projected parameter draws and coerce to draws_matrix (see pa...
Extract projected parameter draws and coerce to matrix
Inverse-link function for augmented-data projection with binomial fami...
Link function for augmented-data projection with binomial family
Augmented-data projection: Internals
Break up matrix terms
Weighted averaging within clusters of parameter draws
Ranking proportions from fold-wise predictor rankings
Run search and performance evaluation with cross-validation
Create cross-validation folds
Execute a function call
Extend a family
Extra family objects
Force search terms
Internal global options
Predictive performance results
Plot ranking proportions from fold-wise predictor rankings
Plot predictive performance
Predictions from a submodel (after projection)
Predictions or log posterior predictive densities from a reference mod...
Predictor terms used in a project() run
Print information about project() output
Print information about a reference model object
Print results (summary) of a varsel() or cv_varsel() run
Print summary of a varsel() or cv_varsel() run
Projection onto submodel(s)
Projection predictive feature selection
Predictor ranking(s)
Reference model and more general information
Create cvfits from cvfun
Retrieve the full-data solution path from a varsel() or `cv_varsel()...
Suggest submodel size
Summary of a varsel() or cv_varsel() run
Run search and performance evaluation without cross-validation
Extract response values, observation weights, and offsets
Performs projection predictive feature selection for generalized linear models (Piironen, Paasiniemi, and Vehtari, 2020, <doi:10.1214/20-EJS1711>) with or without multilevel or additive terms (Catalina, Bürkner, and Vehtari, 2022, <https://proceedings.mlr.press/v151/catalina22a.html>), for some ordinal and nominal regression models (Weber, Glass, and Vehtari, 2025, <doi:10.1007/s00180-024-01506-0>), and for many other regression models (using the latent projection by Catalina, Bürkner, and Vehtari, 2021, <doi:10.48550/arXiv.2109.04702>, which can also be applied to most of the former models). The package is compatible with the 'rstanarm' and 'brms' packages, but other reference models can also be used. See the vignettes and the documentation for more information and examples.
Useful links