Struct ErrorRate 

pub struct ErrorRate {
    pub errors: u64,
    pub total: u64,
}

Errors among trials of a Bernoulli event.

Fields

errors: u64

Number of errors $k$.

total: u64

Number of trials $n$.

Implementations

impl ErrorRate

pub fn add(&mut self, errors: u64, total: u64)

Add errors errors among total new trials.

pub fn merge(&mut self, o: &ErrorRate)

Merge another count.

pub fn rate(&self) -> f64

$k/n$ (NaN without trials).

pub fn wilson(&self, confidence: f64) -> (f64, f64)

Wilson score interval at the given confidence ((0, 1) without trials): eq. (4) of Brown et al. (2001), BinomialInterval::Wilson.

pub fn interval(&self, method: BinomialInterval, confidence: f64) -> (f64, f64)

The interval of the rate by method at the given confidence ((0, 1) without trials; see BinomialInterval: the Wald and Agresti–Coull ends may leave $[0, 1]$, as the paper defines them).

Examples

use comnum_sim::stats::{BinomialInterval, ErrorRate};
let r = ErrorRate { errors: 3, total: 7 };
// Table 5 of Brown et al. (2001): the Jeffreys interval of 3 successes out of 7
let (lo, hi) = r.interval(BinomialInterval::Jeffreys, 0.95);
assert!((lo - 0.139).abs() < 1e-3 && (hi - 0.766).abs() < 1e-3);
// Clopper-Pearson contains Jeffreys, Agresti-Coull contains Wilson
let (cl, ch) = r.interval(BinomialInterval::ClopperPearson, 0.95);
assert!(cl < lo && hi < ch);

Trait Implementations

impl Clone for ErrorRate

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for ErrorRate

impl Debug for ErrorRate

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

Formats the value using the given formatter.

impl Default for ErrorRate

fn default() -> Self

Returns the “default value” for a type.

impl Eq for ErrorRate

impl PartialEq for ErrorRate

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for ErrorRate