Struct PoissonLikelihood 

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

Likelihood for count data under a Poisson model.

Fields

function_name: String

Unique name identifying a likelihood function.

description: String

Free-text description.

formula: String

Mathematical formula.

Implementations

impl PoissonLikelihood

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

Computes the closed-form maximum-likelihood fit λ̂ = x̄ (the sample mean) of the Poisson rate.

The estimate is exact, so the returned MleFit reports converged as true and zero iterations; its AIC/BIC follow the shared information-criteria formulas with k = 1 parameter.

Arguments

• data — the observed counts; each entry must be a finite, non-negative integer, and the sample must contain at least one non-zero count.

Returns

An MleFit whose single parameter is λ̂ and whose log_likelihood is ℓ(λ̂; data).

Errors

• Error::InsufficientData if data is empty.
• Error::InvalidInput if any observation is negative, non-integer, or non-finite.
• Error::DegenerateInput if every observation is zero, which drives λ̂ = 0 and leaves the log-likelihood undefined (ln 0).

Examples

use stats_claw::likelihood::PoissonLikelihood;

let model = PoissonLikelihood::default();
let fit = model.fit(&[2.0, 3.0, 1.0, 5.0, 0.0, 4.0, 2.0, 3.0])?;
// The MLE of the Poisson rate is the sample mean, here 20 / 8 = 2.5.
assert!((fit.params()[0] - 2.5).abs() < 1e-12, "lambda_hat was {}", fit.params()[0]);

Trait Implementations

impl Clone for PoissonLikelihood

fn clone(&self) -> PoissonLikelihood

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PoissonLikelihood

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

Formats the value using the given formatter.

impl Default for PoissonLikelihood

fn default() -> PoissonLikelihood

Returns the “default value” for a type.

impl LogLikelihood for PoissonLikelihood

fn n_params(&self) -> usize

Returns the single free parameter count of the Poisson model, 1 (the rate λ).

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

Evaluates ℓ([λ]; data) = Σᵢ (xᵢ·ln λ − λ − ln Γ(xᵢ + 1)).

Returns f64::NEG_INFINITY when params[0] = λ ≤ 0 (or is NaN), and likewise when any observation is negative, non-integer, or non-finite — all of which lie outside the Poisson support.

Arguments

• params — the one-element rate vector [λ]; a shorter slice yields f64::NEG_INFINITY.
• data — the observed counts.

Returns

The total log-likelihood, or f64::NEG_INFINITY outside the valid domain.