Figure 2.
A generic example of a hierarchical Bayesian model. Hyperparameter values are sampled from the hyperprior (a) and used to parameterize the prior density assigned to the parameter of interest χ (b). In this example, χ follows a lognormal distribution with a mean (μ) and standard deviation (σ). A gamma distribution with shape (s) and rate (β) parameters is the hyperprior and describes the standard deviation hyperparameter (σ) of the lognormal prior on χ.