doi:10.1007/s11222-016-9696-4>. 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.">

loo: Efficient Leave-One-Out Cross-Validation and WAIC for Bayesian Models (original) (raw)

Efficient approximate leave-one-out cross-validation (LOO) for Bayesian models fit using Markov chain Monte Carlo, as described in Vehtari, Gelman, and Gabry (2017) <doi:10.1007/s11222-016-9696-4>. 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.

Version: 2.8.0
Depends: R (≥ 3.1.2)
Imports: checkmate, matrixStats (≥ 0.52), parallel, posterior (≥ 1.5.0), stats
Suggests: bayesplot (≥ 1.7.0), brms (≥ 2.10.0), ggplot2, graphics, knitr, rmarkdown, rstan, rstanarm (≥ 2.19.0), rstantools, spdep, testthat (≥ 2.1.0)
Published: 2024-07-03
DOI: 10.32614/CRAN.package.loo
Author: Aki Vehtari [aut], Jonah Gabry [cre, aut], Måns Magnusson [aut], Yuling Yao [aut], Paul-Christian Bürkner [aut], Topi Paananen [aut], Andrew Gelman [aut], Ben Goodrich [ctb], Juho Piironen [ctb], Bruno Nicenboim [ctb], Leevi Lindgren [ctb]
Maintainer: Jonah Gabry
BugReports: https://github.com/stan-dev/loo/issues
License: GPL (≥ 3)
URL: https://mc-stan.org/loo/, https://discourse.mc-stan.org
NeedsCompilation: no
SystemRequirements: pandoc (>= 1.12.3), pandoc-citeproc
Citation: loo citation info
Materials: NEWS
In views: Bayesian
CRAN checks: loo results

Documentation:

Reference manual: loo.html , <loo.pdf>
Vignettes: Holdout validation and K-fold cross-validation of Stan programs with the loo package (source, R code) Using the loo package (source, R code) Using Leave-one-out cross-validation for large data (source, R code) Approximate leave-future-out cross-validation for Bayesian time series models (source, R code) Mixture IS leave-one-out cross-validation for high-dimensional Bayesian models (source, R code) Avoiding model refits in leave-one-out cross-validation with moment matching (source, R code) Leave-one-out cross-validation for non-factorized models (source, R code) Bayesian Stacking and Pseudo-BMA weights (source, R code) Writing Stan programs for use with the loo package (source, R code)

Downloads:

Reverse dependencies:

Reverse depends: bistablehistory, evidence, spsurv, TriDimRegression
Reverse imports: BAMBI, bayclumpr, bayesdfa, BayesERtools, bayesforecast, BayesGrowth, BayesianFitForecast, bayesnec, beanz, bellreg, blavaan, bmgarch, bmggum, bmscstan, brms, bsitar, causalOT, conformalbayes, disbayes, dynamite, EBcoBART, eDNAjoint, FlexReg, flocker, glmmfields, GPTCM, hbamr, hBayesDM, hdbayes, HeckmanStan, hsstan, LMMELSM, mcmcsae, mcp, measr, MetaStan, missingHE, MixSIAR, multilevelcoda, mvgam, pcFactorStan, phylopairs, projpred, publipha, rater, rbioacc, rmsb, rmstBayespara, rstan, rstanarm, rtmpt, serofoi, shinymrp, StanMoMo, survextrap, tsnet, ubms, vacalibration, walker
Reverse suggests: bayesplot, bayesvl, bmstdr, BSTFA, expertsurv, footBayes, GUD, multinma, neodistr, performance, redist, report, rPBK, sccomp, tipsae, webSDM

Linking:

Please use the canonical formhttps://CRAN.R-project.org/package=looto link to this page.