Implementation of Pareto smoothed importance sampling (PSIS), a method for stabilizing importance ratios. The version of PSIS implemented here corresponds to the algorithm presented in Vehtari, Simpson, Gelman, Yao, and Gabry (2024). For PSIS diagnostics see the pareto-k-diagnostic page.
# S3 method for brmsfit
psis(log_ratios, newdata = NULL, resp = NULL, model_name = NULL, ...)The psis() methods return an object of class "psis",
which is a named list with the following components:
log_weightsVector or matrix of smoothed (and truncated) but unnormalized log
weights. To get normalized weights use the
weights() method provided for objects of
class "psis".
diagnosticsA named list containing two vectors:
pareto_k: Estimates of the shape parameter \(k\) of the
generalized Pareto distribution. See the pareto-k-diagnostic
page for details.
n_eff: PSIS effective sample size estimates.
Objects of class "psis" also have the following attributes:
norm_const_logVector of precomputed values of colLogSumExps(log_weights) that are
used internally by the weights method to normalize the log weights.
tail_lenVector of tail lengths used for fitting the generalized Pareto distribution.
r_effIf specified, the user's r_eff argument.
dimsInteger vector of length 2 containing S (posterior sample size)
and N (number of observations).
methodMethod used for importance sampling, here psis.
A fitted model object of class brmsfit.
Argument is named "log_ratios" to match the argument name of the
loo::psis generic function.
An optional data.frame for which to evaluate predictions. If
NULL (default), the original data of the model is used. NA
values within factors (excluding grouping variables) are interpreted as if
all dummy variables of this factor are zero. This allows, for instance, to
make predictions of the grand mean when using sum coding. NA values
within grouping variables are treated as a new level.
Optional names of response variables. If specified, predictions are performed only for the specified response variables.
Currently ignored.
Further arguments passed to log_lik and
loo::psis.
Vehtari, A., Gelman, A., and Gabry, J. (2017). Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and Computing. 27(5), 1413--1432. doi:10.1007/s11222-016-9696-4 (journal version, preprint arXiv:1507.04544).
Vehtari, A., Simpson, D., Gelman, A., Yao, Y., and Gabry, J. (2024). Pareto smoothed importance sampling. Journal of Machine Learning Research, 25(72):1-58. PDF
if (FALSE) {
fit <- brm(rating ~ treat + period + carry, data = inhaler)
psis(fit)
}
Run the code above in your browser using DataLab