loo (2.1.0)

0 users

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.

Maintainer: Jonah Gabry
Author(s): Aki Vehtari [aut], Andrew Gelman [aut], Jonah Gabry [cre, aut], Yuling Yao [aut], Paul-Christian Brkner [ctb], Ben Goodrich [ctb], Juho Piironen [ctb], Mans Magnusson [ctb]

License: GPL (>= 3)

Uses: checkmate, matrixStats, ggplot2, spdep, testthat, knitr, rmarkdown, brms, rstan, rstanarm, bayesplot, rstantools
Reverse depends: evidence, hBayesDM
Reverse suggests: bayesplot, bayesvl, CopulaDTA, idealstan, performance, psycho, rstan, rstantools, sjstats

Released 9 months ago.

9 previous versions



  (0 votes)


  (0 votes)

Log in to vote.


No one has written a review of loo yet. Want to be the first? Write one now.

Related packages:(20 best matches, based on common tags.)

Search for loo on google, google scholar, r-help, r-devel.

Visit loo on R Graphical Manual.