loo

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

Efficient approximate leave-one-out cross-validation (LOO) for Bayesian models fit using Markov chain Monte Carlo. The approximation uses 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. The package also provides methods for using stacking and other model weighting techniques to average Bayesian predictive distributions.

Total

190,433

Last month

23,467

Last week

5,258

Average per day

782

Daily downloads

Total downloads

Description file content

Package
loo
Type
Package
Title
Efficient Leave-One-Out Cross-Validation and WAIC for Bayesian Models
Version
2.0.0
Date
2018-04-06
Maintainer
Jonah Gabry
URL
BugReports
https://github.com/stan-dev/loo/issues
Description
Efficient approximate leave-one-out cross-validation (LOO) for Bayesian models fit using Markov chain Monte Carlo. The approximation uses 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. The package also provides methods for using stacking and other model weighting techniques to average Bayesian predictive distributions.
License
GPL (>= 3)
LazyData
TRUE
Depends
R (>= 3.1.2)
Imports
graphics, matrixStats (>= 0.52), parallel, stats
Suggests
bayesplot (>= 1.5.0), knitr, rmarkdown, rstan, rstanarm, rstantools, testthat
VignetteBuilder
knitr
Encoding
UTF-8
RoxygenNote
6.0.1
NeedsCompilation
no
Packaged
2018-04-07 02:58:19 UTC; jgabry
Author
Aki Vehtari [aut], Andrew Gelman [aut], Jonah Gabry [cre, aut], Yuling Yao [aut], Juho Piironen [ctb], Ben Goodrich [ctb]
Repository
CRAN
Date/Publication
2018-04-11 12:34:58 UTC

install.packages('loo')

2.0.0

7 months ago

http://mc-stan.org

Jonah Gabry

GPL (>= 3)

Depends on

R (>= 3.1.2)

Imports

graphics, matrixStats (>= 0.52), parallel, stats

Suggests

bayesplot (>= 1.5.0), knitr, rmarkdown, rstan, rstanarm, rstantools, testthat

Discussions