Efficient Leave-One-Out Cross-Validation and WAIC for Bayesian Models

[Up] [Top]

Documentation for package ‘loo’ version 2.7.0

Help Pages

A C E G I K L M N O P R S T U V W

loo-package Efficient LOO-CV and WAIC for Bayesian models

-- A --

ap_psis Pareto smoothed importance sampling (PSIS) using approximate posteriors
ap_psis.array Pareto smoothed importance sampling (PSIS) using approximate posteriors
ap_psis.default Pareto smoothed importance sampling (PSIS) using approximate posteriors
ap_psis.matrix Pareto smoothed importance sampling (PSIS) using approximate posteriors

-- C --

compare Model comparison (deprecated, old version)
crps Continuously ranked probability score
crps.matrix Continuously ranked probability score
crps.numeric Continuously ranked probability score

-- E --

elpd Generic (expected) log-predictive density
elpd.array Generic (expected) log-predictive density
elpd.matrix Generic (expected) log-predictive density
example_loglik_array Objects to use in examples and tests
example_loglik_matrix Objects to use in examples and tests
extract_log_lik Extract pointwise log-likelihood from a Stan model
E_loo Compute weighted expectations
E_loo.default Compute weighted expectations
E_loo.matrix Compute weighted expectations

-- G --

gpdfit Estimate parameters of the Generalized Pareto distribution

-- I --

importance_sampling A parent class for different importance sampling methods.
importance_sampling.array A parent class for different importance sampling methods.
importance_sampling.default A parent class for different importance sampling methods.
importance_sampling.matrix A parent class for different importance sampling methods.
is.kfold Generic function for K-fold cross-validation for developers
is.loo Efficient approximate leave-one-out cross-validation (LOO)
is.psis Pareto smoothed importance sampling (PSIS)
is.psis_loo Efficient approximate leave-one-out cross-validation (LOO)
is.sis Pareto smoothed importance sampling (PSIS)
is.tis Pareto smoothed importance sampling (PSIS)
is.waic Widely applicable information criterion (WAIC)

-- K --

kfold Generic function for K-fold cross-validation for developers
kfold-generic Generic function for K-fold cross-validation for developers
kfold-helpers Helper functions for K-fold cross-validation
kfold_split_grouped Helper functions for K-fold cross-validation
kfold_split_random Helper functions for K-fold cross-validation
kfold_split_stratified Helper functions for K-fold cross-validation
Kline Datasets for loo examples and vignettes

-- L --

loo Efficient approximate leave-one-out cross-validation (LOO)
loo-datasets Datasets for loo examples and vignettes
loo-glossary LOO package glossary
loo.array Efficient approximate leave-one-out cross-validation (LOO)
loo.function Efficient approximate leave-one-out cross-validation (LOO)
loo.matrix Efficient approximate leave-one-out cross-validation (LOO)
loo_approximate_posterior Efficient approximate leave-one-out cross-validation (LOO) for posterior approximations
loo_approximate_posterior.array Efficient approximate leave-one-out cross-validation (LOO) for posterior approximations
loo_approximate_posterior.function Efficient approximate leave-one-out cross-validation (LOO) for posterior approximations
loo_approximate_posterior.matrix Efficient approximate leave-one-out cross-validation (LOO) for posterior approximations
loo_compare Model comparison
loo_compare.default Model comparison
loo_crps Continuously ranked probability score
loo_crps.matrix Continuously ranked probability score
loo_i Efficient approximate leave-one-out cross-validation (LOO)
loo_model_weights Model averaging/weighting via stacking or pseudo-BMA weighting
loo_model_weights.default Model averaging/weighting via stacking or pseudo-BMA weighting
loo_moment_match Moment matching for efficient approximate leave-one-out cross-validation (LOO)
loo_moment_match.default Moment matching for efficient approximate leave-one-out cross-validation (LOO)
loo_moment_match_split Split moment matching for efficient approximate leave-one-out cross-validation (LOO)
loo_predictive_metric Estimate leave-one-out predictive performance..
loo_predictive_metric.matrix Estimate leave-one-out predictive performance..
loo_scrps Continuously ranked probability score
loo_scrps.matrix Continuously ranked probability score
loo_subsample Efficient approximate leave-one-out cross-validation (LOO) using subsampling, so that less costly and more approximate computation is made for all LOO-fold, and more costly and accurate computations are made only for m<N LOO-folds.
loo_subsample.function Efficient approximate leave-one-out cross-validation (LOO) using subsampling, so that less costly and more approximate computation is made for all LOO-fold, and more costly and accurate computations are made only for m<N LOO-folds.

-- M --

mcse_loo Diagnostics for Pareto smoothed importance sampling (PSIS)
milk Datasets for loo examples and vignettes

-- N --

nobs.psis_loo_ss The number of observations in a 'psis_loo_ss' object.

-- O --

obs_idx Get observation indices used in subsampling

-- P --

pareto-k-diagnostic Diagnostics for Pareto smoothed importance sampling (PSIS)
pareto_k_ids Diagnostics for Pareto smoothed importance sampling (PSIS)
pareto_k_influence_values Diagnostics for Pareto smoothed importance sampling (PSIS)
pareto_k_table Diagnostics for Pareto smoothed importance sampling (PSIS)
pareto_k_values Diagnostics for Pareto smoothed importance sampling (PSIS)
plot.loo Diagnostics for Pareto smoothed importance sampling (PSIS)
plot.psis Diagnostics for Pareto smoothed importance sampling (PSIS)
plot.psis_loo Diagnostics for Pareto smoothed importance sampling (PSIS)
pointwise Convenience function for extracting pointwise estimates
pointwise.loo Convenience function for extracting pointwise estimates
print.compare.loo Model comparison
print.compare.loo_ss Model comparison
print.importance_sampling Print methods
print.importance_sampling_loo Print methods
print.loo Print methods
print.psis Print methods
print.psis_loo Print methods
print.psis_loo_ap Print methods
print.waic Print methods
pseudobma_weights Model averaging/weighting via stacking or pseudo-BMA weighting
psis Pareto smoothed importance sampling (PSIS)
psis.array Pareto smoothed importance sampling (PSIS)
psis.default Pareto smoothed importance sampling (PSIS)
psis.matrix Pareto smoothed importance sampling (PSIS)
psislw Pareto smoothed importance sampling (deprecated, old version)
psis_n_eff_values Diagnostics for Pareto smoothed importance sampling (PSIS)

-- R --

relative_eff Convenience function for computing relative efficiencies
relative_eff.array Convenience function for computing relative efficiencies
relative_eff.default Convenience function for computing relative efficiencies
relative_eff.function Convenience function for computing relative efficiencies
relative_eff.importance_sampling Convenience function for computing relative efficiencies
relative_eff.matrix Convenience function for computing relative efficiencies

-- S --

scrps Continuously ranked probability score
scrps.matrix Continuously ranked probability score
scrps.numeric Continuously ranked probability score
sis Standard importance sampling (SIS)
sis.array Standard importance sampling (SIS)
sis.default Standard importance sampling (SIS)
sis.matrix Standard importance sampling (SIS)
stacking_weights Model averaging/weighting via stacking or pseudo-BMA weighting

-- T --

tis Truncated importance sampling (TIS)
tis.array Truncated importance sampling (TIS)
tis.default Truncated importance sampling (TIS)
tis.matrix Truncated importance sampling (TIS)

-- U --

update.psis_loo_ss Update 'psis_loo_ss' objects

-- V --

voice Datasets for loo examples and vignettes
voice_loo Datasets for loo examples and vignettes

-- W --

waic Widely applicable information criterion (WAIC)
waic.array Widely applicable information criterion (WAIC)
waic.function Widely applicable information criterion (WAIC)
waic.matrix Widely applicable information criterion (WAIC)
weights.importance_sampling Extract importance sampling weights