Efficient approximate leave-one-out cross-validation using the loo package

This function uses the loo package to compute PSIS-LOO CV, efficient approximate leave-one-out (LOO) cross-validation for Bayesian models using Pareto smoothed importance sampling (PSIS). This requires that the model was fit using log_lik = TRUE.

Usage

# S3 method for class 'jsdmStanFit'
loo(x, ...)

Arguments

x

The jsdmStanFit model object

...

Other arguments passed to the loo function

Value

A list with class c("psis_loo","loo"), as detailed in the loo documentation