gpdfit

Given a sample \(x\), Estimate the parameters \(k\) and \(\sigma\) of
the generalized Pareto distribution (GPD), assuming the location parameter is
0. By default the fit uses a prior for \(k\), which will stabilize
estimates for very small sample sizes (and low effective sample sizes in the
case of MCMC samples). The weakly informative prior is a Gaussian prior
centered at 0.5.

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

Måns Magnusson

Yuling Yao

Paul-Christian Bürkner

Topi Paananen

Andrew Gelman

Ben Goodrich

Juho Piironen

Bruno Nicenboim

Leevi Lindgren

gpdfit function

<dl><dt>x</dt>
<dd>A numeric vector. The sample from which to estimate the parameters.</dd>
<dt>wip</dt>
<dd>Logical indicating whether to adjust \(k\) based on a weakly
informative Gaussian prior centered on 0.5. Defaults to <code>TRUE</code>.</dd>
<dt>min_grid_pts</dt>
<dd>The minimum number of grid points used in the fitting
algorithm. The actual number used is <code>min_grid_pts + floor(sqrt(length(x)))</code>.</dd>
<dt>sort_x</dt>
<dd>If <code>TRUE</code> (the default), the first step in the fitting
algorithm is to sort the elements of <code>x</code>. If <code>x</code> is already
sorted in ascending order then <code>sort_x</code> can be set to <code>FALSE</code> to
skip the initial sorting step.</dd></dl>

Arguments

Estimate parameters of the Generalized Pareto distribution — gpdfit

<dl>

<dt>x</dt>
<dd>A numeric vector. The sample from which to estimate the parameters.</dd>


<dt>wip</dt>
<dd>Logical indicating whether to adjust \(k\) based on a weakly
informative Gaussian prior centered on 0.5. Defaults to <code>TRUE</code>.</dd>


<dt>min_grid_pts</dt>
<dd>The minimum number of grid points used in the fitting
algorithm. The actual number used is <code>min_grid_pts + floor(sqrt(length(x)))</code>.</dd>


<dt>sort_x</dt>
<dd>If <code>TRUE</code> (the default), the first step in the fitting
algorithm is to sort the elements of <code>x</code>. If <code>x</code> is already
sorted in ascending order then <code>sort_x</code> can be set to <code>FALSE</code> to
skip the initial sorting step.</dd>

</dl>

Estimate parameters of the Generalized Pareto distribution

Join us for
RADAR: AI Edition

gpdfit: Estimate parameters of the Generalized Pareto distribution

Description

Usage

Value

Arguments

Details

References

See Also

Join us for RADAR: AI Edition

Description

Usage

Value

Arguments

Details

References

See Also

Join us for
RADAR: AI Edition