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).

Documentation

Manual: loo.pdf
Vignette: Example

Maintainer: Jonah Gabry <jsg2201 at columbia.edu>

Author(s): Aki Vehtari*, Andrew Gelman*, Jonah Gabry*, Juho Piironen*, Ben Goodrich*

Install package and any missing dependencies by running this line in your R console:

install.packages("loo")

Depends R (>= 3.1.2)
Imports graphics, matrixStats(>=0.50.0), parallel, stats
Suggests knitr, rmarkdown, rstan, rstanarm, testthat
Enhances
Linking to
Reverse
depends
Reverse
imports
beanz, blavaan, brms, ggdmc, hBayesDM, nauf, rstanarm
Reverse
suggests
bayesplot, CopulaDTA, rstan, rstantools
Reverse
enhances
Reverse
linking to

Package loo
Materials
URL http://mc-stan.org/ https://groups.google.com/forum/#!forum/stan-users
Task Views
Version 1.1.0
Published 2017-03-27
License GPL (>= 3)
BugReports https://github.com/stan-dev/loo/issues
SystemRequirements
NeedsCompilation no
Citation
CRAN checks loo check results
Package source loo_1.1.0.tar.gz