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

Efficient approximate leave-one-out cross-validation (LOO) using Pareto smoothed importance sampling (PSIS), a new procedure for regularizing importance weights. As a byproduct of the calculations, we also obtain approximate standard errors for estimated predictive errors and for the comparison of predictive errors between models. We also compute the widely applicable information criterion (WAIC).

Install the latest version of this package by entering the following in R:
install.packages("loo")
AuthorAki Vehtari [aut], Andrew Gelman [aut], Jonah Gabry [cre, aut], Juho Piironen [ctb], Ben Goodrich [ctb]
Date of publication2017-03-27 15:21:52 UTC
MaintainerJonah Gabry <jsg2201@columbia.edu>
LicenseGPL (>= 3)
Version1.1.0
http://mc-stan.org/, https://groups.google.com/forum/#!forum/stan-users

View on CRAN

Files

inst
inst/CITATION
inst/doc
inst/doc/loo-example.R
inst/doc/loo-example.Rmd
inst/doc/loo-example.html
tests
tests/testthat.R
tests/testthat
tests/testthat/E_loo_default_quantile_50.rds
tests/testthat/E_loo_matrix_var.rds
tests/testthat/test_compare.R tests/testthat/test_E_loo.R tests/testthat/test_psislw.R
tests/testthat/E_loo_matrix_quantile_50.rds
tests/testthat/E_loo_default_var.rds
tests/testthat/E_loo_default_quantile_10_90.rds
tests/testthat/E_loo_default_mean.rds
tests/testthat/E_loo_matrix_mean.rds
tests/testthat/E_loo_matrix_quantile_10_90.rds
tests/testthat/test_extract_log_lik.R tests/testthat/test_helpers.R tests/testthat/test_print_plot.R tests/testthat/test_loo_and_waic.R tests/testthat/test_gpdfit.R
NAMESPACE
NEWS.md
R
R/gpdfit.R R/extract_log_lik.R R/helpers.R R/waic.R R/loo_expectation.R R/loo.R R/print.R R/loo_package.R R/psislw.R R/compare.R R/zzz.R R/pareto_k.R
vignettes
vignettes/loo-example.Rmd
MD5
build
build/vignette.rds
DESCRIPTION
man
man/loo.Rd
man/figures
man/figures/stanlogo.png
man/loo-package.Rd man/E_loo.Rd man/psislw.Rd man/waic.Rd man/print.loo.Rd man/nlist.Rd man/compare.Rd man/extract_log_lik.Rd man/gpdfit.Rd man/pareto-k-diagnostic.Rd

Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.