Struct HierarchicalLinearModel 

pub struct HierarchicalLinearModel {
    pub fixed_effects: Array2<f64>,
    pub random_effects_cov: Array2<f64>,
    pub residual_variance: f64,
    pub groups: Array1<usize>,
    pub n_groups: usize,
    pub n_level1_predictors: usize,
    pub n_level2_predictors: usize,
    pub random_slopes: bool,
}

Hierarchical linear model with random intercepts and slopes

Model structure: Level 1: y_ij = β₀j + β₁j * x_ij + ε_ij, ε_ij ~ N(0, σ²) Level 2: β₀j = γ₀₀ + γ₀₁ * w_j + u₀j, u₀j ~ N(0, τ₀₀) β₁j = γ₁₀ + γ₁₁ * w_j + u₁j, u₁j ~ N(0, τ₁₁)

Fields

fixed_effects: Array2<f64>

Fixed effects parameters

random_effects_cov: Array2<f64>

Random effects covariance matrix

residual_variance: f64

Residual variance

groups: Array1<usize>

Group identifiers

n_groups: usize

Number of groups

n_level1_predictors: usize

Number of level-1 predictors

n_level2_predictors: usize

Number of level-2 predictors

random_slopes: bool

Whether to include random slopes

Implementations

impl HierarchicalLinearModel

pub fn new( n_groups: usize, n_level1_predictors: usize, n_level2_predictors: usize, random_slopes: bool, ) -> Result<Self>

Create a new hierarchical linear model

pub fn fit_mcmc<R: Rng + ?Sized>( &mut self, y: ArrayView1<'_, f64>, x_level1: ArrayView2<'_, f64>, x_level2: ArrayView2<'_, f64>, groups: ArrayView1<'_, usize>, n_iter: usize, burnin: usize, rng: &mut R, ) -> Result<HierarchicalModelResults>

Fit the hierarchical model using MCMC

pub fn predict( &self, x_level1: ArrayView2<'_, f64>, x_level2: ArrayView2<'_, f64>, groups: ArrayView1<'_, usize>, ) -> Result<Array1<f64>>

Predict for new data

Trait Implementations

impl Clone for HierarchicalLinearModel

fn clone(&self) -> HierarchicalLinearModel

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for HierarchicalLinearModel

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

Formats the value using the given formatter.