Projection Predictive Feature Selection

---

Documentation for package ‘projpred’ version 2.9.0

• DESCRIPTION file.
• User guides, package vignettes and other documentation.
• Package NEWS.

Help Pages

projpred-package	Projection predictive feature selection
as.matrix.projection	Extract projected parameter draws and coerce to matrix
as_draws.projection	Extract projected parameter draws and coerce to 'draws_matrix' (see package 'posterior')
as_draws_matrix.projection	Extract projected parameter draws and coerce to 'draws_matrix' (see package 'posterior')
augdat_ilink_binom	Inverse-link function for augmented-data projection with binomial family
augdat_link_binom	Link function for augmented-data projection with binomial family
break_up_matrix_term	Break up matrix terms
cl_agg	Weighted averaging within clusters of parameter draws
cv-indices	Create cross-validation folds
cvfolds	Create cross-validation folds
cv_folds	Create cross-validation folds
cv_ids	Create cross-validation folds
cv_proportions	Ranking proportions from fold-wise predictor rankings
cv_proportions.ranking	Ranking proportions from fold-wise predictor rankings
cv_proportions.vsel	Ranking proportions from fold-wise predictor rankings
cv_varsel	Run search and performance evaluation with cross-validation
cv_varsel.default	Run search and performance evaluation with cross-validation
cv_varsel.refmodel	Run search and performance evaluation with cross-validation
cv_varsel.vsel	Run search and performance evaluation with cross-validation
df_binom	Binomial toy example
df_gaussian	Gaussian toy example
extend_family	Extend a family
extra-families	Extra family objects
force_search_terms	Force search terms
get_refmodel	Reference model and more general information
get_refmodel.default	Reference model and more general information
get_refmodel.projection	Reference model and more general information
get_refmodel.refmodel	Reference model and more general information
get_refmodel.stanreg	Reference model and more general information
get_refmodel.vsel	Reference model and more general information
init_refmodel	Reference model and more general information
mesquite	Mesquite data set
performances	Predictive performance results
performances.vsel	Predictive performance results
performances.vselsummary	Predictive performance results
plot.cv_proportions	Plot ranking proportions from fold-wise predictor rankings
plot.ranking	Plot ranking proportions from fold-wise predictor rankings
plot.vsel	Plot predictive performance
pred-projection	Predictions from a submodel (after projection)
predict.refmodel	Predictions or log posterior predictive densities from a reference model
predictor_terms	Predictor terms used in a 'project()' run
predictor_terms.projection	Predictor terms used in a 'project()' run
print.projection	Print information about 'project()' output
print.refmodel	Print information about a reference model object
print.vsel	Print results (summary) of a 'varsel()' or 'cv_varsel()' run
print.vselsummary	Print summary of a 'varsel()' or 'cv_varsel()' run
project	Projection onto submodel(s)
projpred	Projection predictive feature selection
proj_linpred	Predictions from a submodel (after projection)
proj_predict	Predictions from a submodel (after projection)
ranking	Predictor ranking(s)
ranking.vsel	Predictor ranking(s)
refmodel-init-get	Reference model and more general information
run_cvfun	Create 'cvfits' from 'cvfun'
run_cvfun.default	Create 'cvfits' from 'cvfun'
run_cvfun.refmodel	Create 'cvfits' from 'cvfun'
solution_terms	Retrieve the full-data solution path from a 'varsel()' or 'cv_varsel()' run or the predictor combination from a 'project()' run
solution_terms.projection	Retrieve the full-data solution path from a 'varsel()' or 'cv_varsel()' run or the predictor combination from a 'project()' run
solution_terms.vsel	Retrieve the full-data solution path from a 'varsel()' or 'cv_varsel()' run or the predictor combination from a 'project()' run
Student_t	Extra family objects
suggest_size	Suggest submodel size
suggest_size.vsel	Suggest submodel size
summary.vsel	Summary of a 'varsel()' or 'cv_varsel()' run
varsel	Run search and performance evaluation without cross-validation
varsel.default	Run search and performance evaluation without cross-validation
varsel.refmodel	Run search and performance evaluation without cross-validation
varsel.vsel	Run search and performance evaluation without cross-validation
y_wobs_offs	Extract response values, observation weights, and offsets