neighbayes.models.priors.MultilevelPriors

class neighbayes.models.priors.MultilevelPriors(beta_mu=None, beta_sigma=None, sigma2_alpha=2.0, sigma2_beta=None, nu=4.0, rho_lower=None, rho_upper=None, sigma_nu=3.0, sigma_scale=None, level_beta_mu=None, level_beta_sigma=None)[source]

Priors for neighbayes.models.Multilevel.

beta_mu, beta_sigma

Normal prior on the units’ coefficients \(\beta_0\); default Gelman et al. (2008).

sigma2_alpha, sigma2_beta

\(\sigma_0^2 \sim \text{Inv-}\Gamma\), the data-informed prior of the single-level models (LeSage). sigma2_beta defaults to the pooled variance of y within level-1 groups, the units’ share of Var(y): Var(y) itself also counts the upper levels’ variance, and the prior shifts the posterior mean of \(\sigma_0^2\) by about 2 * sigma2_beta / n, which with strong upper levels was several posterior standard deviations.

rho_lower, rho_upper

Uniform bounds on every level’s autoregressive parameter (ρ or λ); default each graph’s stability bounds, (-1, 1) when row-standardized.

sigma_nu, sigma_scale

Half-t prior on each upper level’s innovation sd, \(\sigma_\ell \sim \text{half-}t_\nu(0, A)\). sigma_scale (A) defaults to sd(y): every level’s effects are in the outcome’s units. nu = 3 regularizes more than the half-Cauchy when a level has few groups (a few dozen states, say). NUTS and Gibbs use the same prior.

level_beta_mu, level_beta_sigma

Overrides of the Normal prior on an upper level’s coefficients, keyed by level number or name, e.g. {"county": 1.0}. The default is Gelman et al. (2008) with the outcome’s scale and each level’s covariates.

__init__(beta_mu=None, beta_sigma=None, sigma2_alpha=2.0, sigma2_beta=None, nu=4.0, rho_lower=None, rho_upper=None, sigma_nu=3.0, sigma_scale=None, level_beta_mu=None, level_beta_sigma=None)[source]

Methods

__init__([beta_mu, beta_sigma, ...])

Attributes

beta_mu

beta_sigma

level_beta_mu

level_beta_sigma

nu

rho_lower

rho_upper

sigma2_alpha

sigma2_beta

sigma_nu

sigma_scale

beta_mu = None[source]
beta_sigma = None[source]
level_beta_mu = None[source]
level_beta_sigma = None[source]
nu = 4.0[source]
rho_lower = None[source]
rho_upper = None[source]
sigma2_alpha = 2.0[source]
sigma2_beta = None[source]
sigma_nu = 3.0[source]
sigma_scale = None[source]