Struct DenseFlmmResult 

#[non_exhaustive]
pub struct DenseFlmmResult {

    pub mean_function: Vec<f64>,
    pub beta_functions: FdMatrix,
    pub random_effects: FdMatrix,
    pub fitted: FdMatrix,
    pub residuals: FdMatrix,
    pub random_variance: Vec<f64>,
    pub sigma2_eps: f64,
    pub sigma2_u: Vec<f64>,
    pub sigma2_slope: Vec<f64>,
    pub ncomp: usize,
    pub n_subjects: usize,
    pub eigenvalues: Vec<f64>,
    pub n_iter: usize,
    pub converged: bool,
}

Result of a dense functional linear mixed model fit.

Parametrization note

fdars formulates the model over FPC scores (reusing fdata_to_pc) rather than over spline/basis coefficients as in R’s denseFLMM package. Consequently variance components are per FPC component, not smoothed over the argument domain.

Random-slope estimation is not implemented in this release; sigma2_slope is always zero-filled (one entry per FPC component). Future releases may add two-random-effect scalar LMM support via a dedicated helper.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

mean_function: Vec<f64>

Overall mean function μ̂(t) (length m)

beta_functions: FdMatrix

Fixed effect coefficient functions β̂_j(t) (p × m matrix, one row per covariate)

random_effects: FdMatrix

Random effect functions b̂_i(t) per subject (n_subjects × m)

fitted: FdMatrix

Fitted values for all observations (n_total × m)

residuals: FdMatrix

Residuals (n_total × m)

random_variance: Vec<f64>

Variance of random effects at each time point Var_i(b̂_i(t)) (length m)

sigma2_eps: f64

Mean residual variance averaged across FPC-score component models.

Each per-component model operates on L²-normalized scores; this average is on the normalized scale and is not directly comparable to the marginal residual variance σ²_ε from R’s lmer(). See the struct-level parametrization note.

sigma2_u: Vec<f64>

Random-intercept variance per FPC component (length k)

sigma2_slope: Vec<f64>

Random-slope variance per FPC component (zero-filled when random_slopes = false)

ncomp: usize

Number of FPC components actually used

n_subjects: usize

Number of unique subjects

eigenvalues: Vec<f64>

FPC eigenvalues (singular values squared / n_total), length k

n_iter: usize

Maximum number of REML EM iterations reached across components

converged: bool

true if all component models converged before config.max_iter

Trait Implementations

impl Clone for DenseFlmmResult

fn clone(&self) -> DenseFlmmResult

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for DenseFlmmResult

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

Formats the value using the given formatter.

impl PartialEq for DenseFlmmResult

fn eq(&self, other: &DenseFlmmResult) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for DenseFlmmResult