Struct Normal 

pub struct Normal {
    pub mean: f64,
    pub std_dev: f64,
}

Normal (Gaussian) distribution N(μ, σ²) as a typed struct.

Implements Distribution for use with fit_all / fit_best.

Examples

use rs_stats::distributions::normal_distribution::Normal;
use rs_stats::distributions::traits::Distribution;

let n = Normal::new(0.0, 1.0).unwrap();
assert!((n.mean() - 0.0).abs() < 1e-10);
assert!((n.pdf(0.0).unwrap() - 0.398_942_280_401_4).abs() < 1e-10);

Fields

mean: f64

Mean μ

std_dev: f64

Standard deviation σ (must be > 0)

Implementations

impl Normal

pub fn new(mean: f64, std_dev: f64) -> StatsResult<Self>

Creates a Normal distribution with validation.

pub fn fit(data: &[f64]) -> StatsResult<Self>

Maximum-likelihood estimate from data.

MLE: μ = mean(data), σ = population std-dev. Single-pass online (Welford) — never walks data twice and never allocates.

Trait Implementations

impl Clone for Normal

fn clone(&self) -> Normal

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Normal

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

Formats the value using the given formatter.

impl Distribution for Normal

fn name(&self) -> &str

Human-readable distribution name, e.g. "Normal".

fn num_params(&self) -> usize

Number of free parameters (used when computing AIC / BIC).

fn pdf(&self, x: f64) -> StatsResult<f64>

Probability density function f(x).

fn logpdf(&self, x: f64) -> StatsResult<f64>

Natural logarithm of the PDF: ln f(x).

fn cdf(&self, x: f64) -> StatsResult<f64>

Cumulative distribution function F(x) = P(X ≤ x).

fn inverse_cdf(&self, p: f64) -> StatsResult<f64>

Quantile (inverse CDF): find x such that F(x) = p.

fn mean(&self) -> f64

Mean (expected value) μ.

fn variance(&self) -> f64

Variance σ².

fn std_dev(&self) -> f64

Standard deviation σ = √(variance).

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

Sum of log-likelihoods: Σ ln f(xᵢ).

fn aic(&self, data: &[f64]) -> StatsResult<f64>

Akaike Information Criterion: AIC = 2k − 2·ln(L̂).

fn bic(&self, data: &[f64]) -> StatsResult<f64>

Bayesian Information Criterion: BIC = k·ln(n) − 2·ln(L̂).

impl Copy for Normal