Balík: r-cran-loo (2.0.0-2)

[ Zdroj: r-cran-loo  ]

GNU R 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, it is possible as well to 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.

Ostatné balíky súvisiace s balíkom r-cran-loo

• dep: r-api-3.5

  virtuálny balík poskytovaný balíkom r-base-core

• dep: r-base-core (>= 3.5.0-5)

  jadro GNU R na štatistické výpočty a grafický systém

• dep: r-cran-matrixstats (>= 0.52)

  GNU R methods that apply to rows and columns of a matrix

• sug: libgdal-dev

  Geospatial Data Abstraction Library - vývojové súbory

• sug: r-cran-bayesplot (>= 1.5.0)

  GNU R plotting for bayesian models

• sug: r-cran-knitr

  balík GNU R na dynamickú tvorbu zostáv pomocou Literate Programming

• sug: r-cran-rmarkdown

  convert R markdown documents into a variety of formats

• sug: r-cran-rstan

  GNU R interface to Stan

• sug: r-cran-rstantools

  tools for developing GNU R packages interfacing with 'Stan'

• sug: r-cran-testthat

  GNU R testsuite

Stiahnuť r-cran-loo

Stiahnuť pre všetky dostupné architektúry

Architektúra	Veľkosť balíka	Nainštalovaná veľkosť	Súbory
all	748.1 kB	1,064.0 kB	[zoznam súborov]