Posterior KDEs

KDE plot of the variable mu from the centered eight model. sample_dims restricts the KDE computation to the draw dimension only.

using ArviZPythonPlots, ArviZExampleData

use_style("arviz-variat")

data = load_example_data("centered_eight")
pc = plot_dist(data; kind="kde", var_names=["mu"], sample_dims=["draw"])
pc.add_title("KDE of μ by Chain (Centered Eight)")

See plot_dist.

See also the EABM chapter on Visualization of Random Variables with ArviZ.