Compare DPMirt models using information criteria

Compare DPMirt models using information criteria

Usage

dpmirt_compare(..., criterion = "waic")

Arguments

...

Two or more dpmirt_fit objects.

criterion

Character. Comparison criterion. Currently only "waic" is supported.

Value

A data.frame ranking models by the criterion with columns model, waic, delta_waic, and waic_aggregation. Fits with unavailable WAIC are kept in the table with NA deltas; all-missing WAIC values raise an error.

Details

For chain-labeled multi-run fits, waic_aggregation records whether the top-level WAIC is a mean of per-run WAIC values. Such values are retained for provenance and backward compatibility but should not be interpreted as pooled posterior WAIC.

See also

dpmirt, dpmirt_diagnostics

Other diagnostics: dpmirt_diagnostics()

Examples

if (FALSE) { # \dontrun{
sim <- dpmirt_simulate(200, 20, model = "rasch", seed = 42)
fit1 <- dpmirt(sim$response, model = "rasch", prior = "normal",
               niter = 5000, nburnin = 1000, seed = 123)
fit2 <- dpmirt(sim$response, model = "rasch", prior = "dpm",
               niter = 5000, nburnin = 1000, seed = 123)

# Compare via WAIC
comp <- dpmirt_compare(fit1, fit2)
print(comp)
} # }