Function consistency 

pub fn consistency(
    prior: &Prior,
    estimate: &[f64],
    posterior: &FlatMatrix,
) -> Result<Option<Consistency>, FittingError>

The consistency of the fitted values estimate of a prior’s parameters with it, in the prior’s order, given the fit’s covariance posterior of them.

With C = LLᵀ the prior covariance, z = L⁻¹(estimate − mean) and I − L⁻¹·posterior·L⁻ᵀ = Σᵢ rᵢuᵢuᵢᵀ, rᵢ is the share of the prior’s variance along uᵢ that the counts remove, and q = Σᵢ (uᵢᵀz)²/rᵢ over the dof directions with rᵢ ≥ 8e-4. If the prior and the counts’ model are right and estimate and posterior are the fit’s without its bounds (Unbounded), q follows χ² with dof degrees of freedom, and p = P(χ²_dof ≥ q).

None when posterior is not finite or no direction has rᵢ ≥ 8e-4.

Errors

FittingError::LengthMismatch if estimate or posterior does not match the prior’s parameters; FittingError::InvalidConfig if the prior’s mean is not finite or a measured sd not finite and positive, estimate is not finite while posterior is, or posterior is not symmetric to 1e-12 of √(Σᵢᵢ Σⱼⱼ) or not within the prior, with some rᵢ below −8e-4 or above 1 + 8e-4, or q is not a number, as overflow in whitening can give; FittingError::EvaluationFailed if a decomposition fails.