Struct ExponentialLikelihood 

pub struct ExponentialLikelihood {
    pub function_name: String,
    pub description: String,
    pub formula: String,
}

Likelihood for waiting-time data under an exponential model.

Fields

function_name: String

Unique name identifying a likelihood function.

description: String

Free-text description.

formula: String

Mathematical formula.

Implementations

impl ExponentialLikelihood

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

Fits this exponential model to data by its closed-form maximum-likelihood estimate λ̂ = n / Σxᵢ.

The rate MLE is analytic, so the returned MleFit reports converged = true and zero iterations; its log_likelihood is ℓ(λ̂; data) and the AIC/BIC follow from k = 1 parameter and n = data.len().

Arguments

• data — the observed sample; every value must be ≥ 0 (the exponential support) and the sample must be non-empty with a strictly positive sum.

Returns

An MleFit whose single parameter is the rate estimate λ̂.

Errors

• Error::InsufficientData if data is empty.
• Error::InvalidInput if any observation is negative (outside the exponential support) or non-finite.
• Error::DegenerateInput if every observation is zero (Σxᵢ = 0), which would make λ̂ = n / 0 infinite.

Examples

use stats_claw::likelihood::ExponentialLikelihood;

let fit = ExponentialLikelihood::default().fit(&[1.0, 2.0, 3.0])?;
// λ̂ = 3 / 6 = 0.5.
assert!((fit.params()[0] - 0.5).abs() < 1e-12, "lambda_hat was {}", fit.params()[0]);
assert!(fit.converged());

Trait Implementations

impl Clone for ExponentialLikelihood

fn clone(&self) -> ExponentialLikelihood

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ExponentialLikelihood

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

Formats the value using the given formatter.

impl Default for ExponentialLikelihood

fn default() -> ExponentialLikelihood

Returns the “default value” for a type.

impl LogLikelihood for ExponentialLikelihood

fn n_params(&self) -> usize

Returns 1: the exponential rate model has the single parameter λ.

Examples

use stats_claw::likelihood::ExponentialLikelihood;
use stats_claw::likelihood::LogLikelihood;

assert_eq!(ExponentialLikelihood::default().n_params(), 1);

fn log_likelihood(&self, params: &[f64], data: &[f64]) -> f64

Evaluates the total exponential log-likelihood ℓ(λ; data) = n·ln λ − λ·Σxᵢ.

Arguments

• params — the one-element rate vector [λ]; only params[0] is read.
• data — the observed sample.

Returns

The scalar log-likelihood, or f64::NEG_INFINITY when λ ≤ 0, or any observation is negative or non-finite — all lie outside the model’s valid domain. Per the LogLikelihood contract a non-finite observation (NaN or ±∞) yields −∞ rather than propagating a NaN/incidental −∞.

Examples

use stats_claw::likelihood::ExponentialLikelihood;
use stats_claw::likelihood::LogLikelihood;

let model = ExponentialLikelihood::default();
// ℓ(1; [1, 2]) = 2·ln 1 − 1·3 = −3.
assert!((model.log_likelihood(&[1.0], &[1.0, 2.0]) + 3.0).abs() < 1e-12);
// A non-positive rate is outside the domain.
assert_eq!(model.log_likelihood(&[0.0], &[1.0]), f64::NEG_INFINITY);