$ ⌘K
// constructs / posterior_predictive_check

Posterior Predictive Check

id: posterior_predictive_check

A model-adequacy diagnostic that simulates replicated datasets from the posterior predictive distribution and compares summary statistics (mean, variance, quantiles, test quantities) of the replications against the observed data. Systematic discrepancies indicate model misspecification. The standard goodness-of-fit procedure in the Bayesian workflow.

// usage

Used in 1 pax

↑ positive relationship 2 findings