weights.importance_sampling

Extract importance sampling weights

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.

Jonah Gabry

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

Aki Vehtari

Mans Magnusson

Yuling Yao

Paul-Christian Bürkner

Topi Paananen

Andrew Gelman

Ben Goodrich

Juho Piironen

Bruno Nicenboim

Leevi Lindgren

weights.importance_sampling function

<dl><dt>object</dt>
<dd>An object returned by <code>psis()</code>, <code>tis()</code>, or <code>sis()</code>.</dd>
<dt>...</dt>
<dd>Ignored.</dd>
<dt>log</dt>
<dd>Should the weights be returned on the log scale? Defaults to
<code>TRUE</code>.</dd>
<dt>normalize</dt>
<dd>Should the weights be normalized? Defaults to <code>TRUE</code>.</dd></dl>

Arguments

Extract importance sampling weights — weights.importance_sampling

<dl>

<dt>object</dt>
<dd>An object returned by <code>psis()</code>, <code>tis()</code>, or <code>sis()</code>.</dd>


<dt>...</dt>
<dd>Ignored.</dd>


<dt>log</dt>
<dd>Should the weights be returned on the log scale? Defaults to
<code>TRUE</code>.</dd>


<dt>normalize</dt>
<dd>Should the weights be normalized? Defaults to <code>TRUE</code>.</dd>

</dl>

Extract importance sampling weights

weights.importance_sampling: Extract importance sampling weights

Description

Usage

Value

Arguments

Examples