LOO PAV-adjusted calibration

PAV-adjusted calibration plot using leave-one-out (LOO) cross-validation to resample the posterior predictive distribution. Ideal for binary, ordinal or categorical data.

Matplotlib

Bokeh

Link to this page with the bokeh tab selected

Plotly

Link to this page with the plotly tab selected

from arviz_base import load_arviz_data

import arviz_plots as azp

azp.style.use("arviz-variat")

dt = load_arviz_data("anes")
pc = azp.plot_loo_pava(
    dt,
    backend="none",
)
pc.show()

See also

API Documentation: plot_loo_pava

EABM chapter on Posterior predictive checks for binary data

Other examples with plot_loo_pava

Download Python Source Code: plot_loo_pava.py