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).

AuthorAki Vehtari [aut], Andrew Gelman [aut], Jonah Gabry [cre, aut], Juho Piironen [ctb], Ben Goodrich [ctb]
Date of publication2016-12-16 08:33:56
MaintainerJonah Gabry <jsg2201@columbia.edu>
LicenseGPL (>= 3)
Version1.0.0
http://mc-stan.org/, https://groups.google.com/forum/#!forum/stan-users

View on CRAN

Files

loo
loo/inst
loo/inst/CITATION
loo/inst/doc
loo/inst/doc/loo-example.R
loo/inst/doc/loo-example.Rmd
loo/inst/doc/loo-example.html
loo/tests
loo/tests/testthat.R
loo/tests/testthat
loo/tests/testthat/test_compare.R
loo/tests/testthat/test_psislw.R
loo/tests/testthat/test_extract_log_lik.R
loo/tests/testthat/test_helpers.R
loo/tests/testthat/test_print_plot.R
loo/tests/testthat/test_loo_and_waic.R
loo/tests/testthat/test_gpdfit.R
loo/NAMESPACE
loo/NEWS.md
loo/R
loo/R/gpdfit.R loo/R/extract_log_lik.R loo/R/helpers.R loo/R/waic.R loo/R/loo.R loo/R/print.R loo/R/loo_package.R loo/R/psislw.R loo/R/compare.R loo/R/zzz.R loo/R/pareto_k.R
loo/vignettes
loo/vignettes/loo-example.Rmd
loo/MD5
loo/build
loo/build/vignette.rds
loo/DESCRIPTION
loo/man
loo/man/loo.Rd
loo/man/figures
loo/man/figures/stanlogo.png
loo/man/loo-package.Rd loo/man/psislw.Rd loo/man/waic.Rd loo/man/print.loo.Rd loo/man/nlist.Rd loo/man/compare.Rd loo/man/extract_log_lik.Rd loo/man/gpdfit.Rd loo/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.