Struct BmsLogslopeJacobian 

pub struct BmsLogslopeJacobian {
    pub marginal_dense: Arc<Array2<f64>>,
    pub logslope_dense: Arc<Array2<f64>>,
    pub offset_m: Array1<f64>,
    pub offset_s: Array1<f64>,
    pub z: Arc<Array1<f64>>,
    pub p_marginal: usize,
}

β-dependent Jacobian for the BMS logslope block.

∂η_i/∂β_s = (q_i · s²·g_i / c_i + s·z_i) · G[i,:] where q_i = M[i,:] · β_m + offset_m[i], g_i = G[i,:] · β_s + offset_s[i], c_i = sqrt(1 + (s·g_i)²), s = state.probit_frailty_scale.

probit_frailty_scale is read from the evaluation state at call time.

Designs are pre-densified at construction to avoid repeated materialisation.

Fields

marginal_dense: Arc<Array2<f64>>

Dense marginal design: n × p_m.

logslope_dense: Arc<Array2<f64>>

Dense logslope design: n × p_s.

offset_m: Array1<f64>

offset_s: Array1<f64>

z: Arc<Array1<f64>>

p_marginal: usize

Number of marginal columns (= start of β_s in the full β vector).

Implementations

impl BmsLogslopeJacobian

pub fn new( marginal_dense: Arc<Array2<f64>>, logslope_dense: Arc<Array2<f64>>, offset_m: Array1<f64>, offset_s: Array1<f64>, z: Arc<Array1<f64>>, p_marginal: usize, ) -> Self

Trait Implementations

impl BlockEffectiveJacobian for BmsLogslopeJacobian

fn effective_jacobian_rows( &self, state: &FamilyLinearizationState<'_>, rows: Range<usize>, ) -> Result<Array2<f64>, String>

Stacked multi-output Jacobian for a contiguous observation row range.

fn n_outputs(&self) -> usize

Number of stacked output channels. 1 for most blocks.

fn effective_jacobian_at( &self, state: &FamilyLinearizationState<'_>, ) -> Result<ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>, String>

Stacked multi-output Jacobian at the current β.

fn eta_row_scaling_for_skewness(&self) -> Option<Arc<[f64]>>

Returns the per-row scaling vector when this callback is a simple diagonal-scaling block (RowScaledJacobian). Used by the identifiability audit’s skewness-aware bias correction (T25).