Struct RhoPosteriorMixture 

pub struct RhoPosteriorMixture {
    pub nodes: Vec<RhoMixtureNode>,
    pub mean: Array1<f64>,
    pub covariance: Array2<f64>,
    pub effective_sample_size: f64,
}

Tier-1 deliverable (#938): π(ρ|y) as a discrete mixture of conditional Gaussians, with the posterior moment summary of ρ itself.

The conditional Gaussian at each node is exactly what the engine already produces at fixed ρ; this struct owns the node locations and weights, and mixture_coefficient_covariance (monolith inference::rho_posterior) assembles the mixture-corrected coefficient covariance from per-node conditionals supplied by the caller.

Fields

nodes: Vec<RhoMixtureNode>

Quadrature nodes with normalized weights (weights sum to 1).

mean: Array1<f64>

Posterior mean of ρ: Σ_m w_m ρ_m.

covariance: Array2<f64>

Posterior covariance of ρ: Σ_m w_m (ρ_m−ρ̄)(ρ_m−ρ̄)ᵀ.

effective_sample_size: f64

Kish ESS of the node weights (Σw)²/Σw² — how non-Gaussian the exact posterior is relative to the Laplace proposal (max = node count).

Trait Implementations

impl Clone for RhoPosteriorMixture

fn clone(&self) -> RhoPosteriorMixture

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for RhoPosteriorMixture

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.