Struct SmoothTestInput 

pub struct SmoothTestInput<'a> {
    pub beta: ArrayBase<ViewRepr<&'a f64>, Dim<[usize; 1]>>,
    pub covariance: &'a ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>,
    pub influence_matrix: Option<&'a ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>>,
    pub whitening_gram: Option<&'a ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>>,
    pub coeff_range: Range<usize>,
    pub edf: f64,
    pub nullspace_dim: usize,
    pub residual_df: Option<f64>,
    pub scale: SmoothTestScale,
}

Inputs to wood_smooth_test. beta is the full coefficient vector; the term block being tested is beta[coeff_range]. covariance is the matching posterior covariance Σ̂ (full p×p; the diagonal block is sliced out). covariance must be the scale-included posterior covariance (mgcv Vb/Vp, i.e. H⁻¹ already multiplied by the dispersion φ̂), so the Wald statistic T = β̂'·Σ̂⁻·β̂ is dimensionless — the residual dispersion has already been divided out and the F-statistic is T/ref_df with no further φ̂ factor. influence_matrix is the optional coefficient-space influence F = H⁻¹ X'WX; when present tr(F_jj)² / tr(F_jj²) is used as the Wood-corrected reference d.f. whitening_gram is the optional term-block-aligned weighted design Gram G = X'WX (H − S(λ), full p×p, same coefficient layout as covariance); when present the covariance is mapped into the Wood (2013) fitted-value space R·V·Rᵀ (RᵀR = G) before the rank-r truncation, so the pseudo-inverse keeps the directions that carry the estimated function rather than the raw largest-variance (heavily-penalized) coefficient directions. When absent the raw coefficient covariance is truncated directly — a graceful fallback for persisted models whose Gram was not serialized. edf is the smooth’s effective d.f. (rank of the truncated pseudo-inverse); nullspace_dim is the fixed-effect (unpenalized) leading dimension within the block, used as a floor on the truncation rank (those directions are never shrunk and must always be tested). residual_df is the denominator d.f. for the Estimated-scale F branch. It is None when that inference geometry is unavailable; the estimated-scale test then returns None instead of inventing denominator degrees of freedom. The known-scale branch does not consume it.

Fields

beta: ArrayBase<ViewRepr<&'a f64>, Dim<[usize; 1]>>

covariance: &'a ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>

influence_matrix: Option<&'a ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>>

whitening_gram: Option<&'a ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>>

coeff_range: Range<usize>

edf: f64

nullspace_dim: usize

residual_df: Option<f64>

scale: SmoothTestScale

Trait Implementations

impl<'a> Clone for SmoothTestInput<'a>

fn clone(&self) -> SmoothTestInput<'a>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<'a> Debug for SmoothTestInput<'a>

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

Formats the value using the given formatter.