Struct BinomialLikelihood 

pub struct BinomialLikelihood {
    pub number_of_trials: i64,
    pub function_name: String,
    pub description: String,
    pub formula: String,
}

Likelihood for success counts under a binomial model.

Fields

number_of_trials: i64

Number of trials.

function_name: String

Unique name identifying a likelihood function.

description: String

Free-text description.

formula: String

Mathematical formula.

Implementations

impl BinomialLikelihood

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

Closed-form maximum-likelihood fit of the success probability p.

The binomial MLE is p̂ = (Σᵢ xᵢ) / (N · n_trials) — the pooled success rate across all N draws. Because the estimate is analytic, the returned MleFit reports converged() == true and iterations() == 0, with the AIC/BIC computed for k = 1 parameter over N observations.

Arguments

• data — the observed per-draw success counts; each must be an integer in [0, n_trials]. Must be non-empty.

Returns

An MleFit whose single parameter is p̂.

Errors

• Error::InsufficientData if data is empty.
• Error::InvalidInput if this model’s number_of_trials is not a positive integer, or if any observation is negative, non-integral, or greater than number_of_trials.
• Error::DegenerateInput if every observation is 0 or every observation is n_trials (so p̂ lands on the boundary {0, 1} and the log-likelihood is degenerate).

Examples

use stats_claw::likelihood::BinomialLikelihood;

let model = BinomialLikelihood { number_of_trials: 10, ..Default::default() };
let fit = model.fit(&[3.0, 5.0, 2.0, 4.0, 6.0])?;
// p̂ = 20 / (5·10) = 0.4.
assert!((fit.params()[0] - 0.4).abs() < 1e-12, "p_hat was {}", fit.params()[0]);
assert!(fit.converged());

Trait Implementations

impl Clone for BinomialLikelihood

fn clone(&self) -> BinomialLikelihood

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BinomialLikelihood

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

Formats the value using the given formatter.

impl Default for BinomialLikelihood

fn default() -> BinomialLikelihood

Returns the “default value” for a type.

impl LogLikelihood for BinomialLikelihood

fn n_params(&self) -> usize

The binomial has a single free parameter, the success probability p.

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

Evaluates ℓ(p; data) for the success probability params[0].

Returns f64::NEG_INFINITY outside the valid domain: when number_of_trials is not positive, p ∉ (0, 1), or any observation is negative, non-integral, or exceeds number_of_trials.