loo (version 2.8.0)

elpd: Generic (expected) log-predictive density

Description

The elpd() methods for arrays and matrices can compute the expected log pointwise predictive density for a new dataset or the log pointwise predictive density of the observed data (an overestimate of the elpd).

Usage

elpd(x, ...)
# S3 method for array
elpd(x, ...)
# S3 method for matrix
elpd(x, ...)

Arguments

x

A log-likelihood array or matrix. The Methods (by class) section, below, has detailed descriptions of how to specify the inputs for each method.

...

Currently ignored.

Methods (by class)

• elpd(array): An $I$ by $C$ by $N$ array, where $I$ is the number of MCMC iterations per chain, $C$ is the number of chains, and $N$ is the number of data points.

• elpd(matrix): An $S$ by $N$ matrix, where $S$ is the size of the posterior sample (with all chains merged) and $N$ is the number of data points.

Details

The elpd() function is an S3 generic and methods are provided for 3-D pointwise log-likelihood arrays and matrices.

See Also

The vignette Holdout validation and K-fold cross-validation of Stan programs with the loo package for demonstrations of using the elpd() methods.

Examples

# Calculate the lpd of the observed data
LLarr <- example_loglik_array()
elpd(LLarr)