Struct MaximumLikelihood 

pub struct MaximumLikelihood {
    pub estimation_method: String,
    pub convergence_tolerance: f64,
    pub function_name: String,
    pub description: String,
    pub formula: String,
}

General maximum-likelihood estimation procedure.

Fields

estimation_method: String

Estimation method.

convergence_tolerance: f64

Convergence tolerance.

function_name: String

Unique name identifying a likelihood function.

description: String

Free-text description.

formula: String

Mathematical formula.

Implementations

impl MaximumLikelihood

pub fn fit( &self, model: &impl LogLikelihood, data: &[f64], init: &[f64], ) -> Result<MleFit>

Fits model to data from init, using this instance’s convergence_tolerance.

The stored tolerance is the gradient-norm threshold forwarded to fit_mle. When the field is left at its default of 0.0 (a non-positive, unusable threshold), it falls back to 1e-8.

Arguments

• model — the parametric log-likelihood to fit.
• data — the observed sample; must be non-empty.
• init — the starting parameter vector; length must equal model.n_params().

Errors

Propagates the errors of fit_mle: Error::InsufficientData for empty data, and Error::InvalidInput for an init/parameter-count mismatch, a non-positive resolved tolerance, or an init outside the model’s valid domain (non-finite log-likelihood).

Examples

use stats_claw::likelihood::MaximumLikelihood;
use stats_claw::likelihood::LogLikelihood;

struct MeanModel;
impl LogLikelihood for MeanModel {
    fn n_params(&self) -> usize { 1 }
    fn log_likelihood(&self, p: &[f64], d: &[f64]) -> f64 { -d.iter().map(|x| (x - p[0]).powi(2)).sum::<f64>() }
}

let mle = MaximumLikelihood { convergence_tolerance: 1e-9, ..Default::default() };
let fit = mle.fit(&MeanModel, &[2.0, 4.0, 6.0], &[0.0])?;
assert!((fit.params()[0] - 4.0).abs() < 1e-5, "mu_hat was {}", fit.params()[0]);

Trait Implementations

impl Clone for MaximumLikelihood

fn clone(&self) -> MaximumLikelihood

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MaximumLikelihood

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

Formats the value using the given formatter.

impl Default for MaximumLikelihood

fn default() -> MaximumLikelihood

Returns the “default value” for a type.