Struct WaveletLeaderScaling 

pub struct WaveletLeaderScaling {
    pub moments: Vec<f64>,
    pub scales: Vec<usize>,
    pub structure_functions: Vec<Vec<f64>>,
    pub scaling_exponents: Vec<f64>,
}

Raw wavelet-leader scaling data.

This exposes the intermediate quantities used by the WLMF method: structure functions S(j, q) and scaling exponents ζ(q) obtained from linear regressions of ln S(j, q) against scale j.

Fields

moments: Vec<f64>

Moments q used in the analysis.

scales: Vec<usize>

Scales j used (decomposition levels).

structure_functions: Vec<Vec<f64>>

Structure functions S(j, q) for each moment. Outer index: moment index, inner index: scale index.

scaling_exponents: Vec<f64>

Scaling exponents ζ(q) for each moment q.

Implementations

impl WaveletLeaderScaling

pub fn multifractality_score(&self) -> Option<f64>

Compute a simple multifractality score based on curvature of ζ(q) near q = 0.

This uses a discrete second difference around q = 0:

ζ’’(0) ≈ ζ(1) − 2ζ(0) + ζ(−1)

and returns |ζ''(0)|. For monofractal processes, ζ(q) is approximately linear in q so this score is close to 0, while strongly multifractal processes exhibit larger curvature and thus larger scores.

Returns None if the moment set does not contain q = −1, 0, 1.

pub fn linearity_r2(&self) -> Option<f64>

Compute the R² of the best linear fit ζ(q) ≈ a·q + b.

This provides a simple monofractality score: values close to 1.0 indicate that ζ(q) is well-approximated by a straight line in q, while significantly smaller values indicate nonlinear scaling (and hence multifractality).

Returns None if fewer than two distinct q-values are available.

Trait Implementations

impl Clone for WaveletLeaderScaling

fn clone(&self) -> WaveletLeaderScaling

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for WaveletLeaderScaling

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

Formats the value using the given formatter.