Struct stats_ci::proportion::Stats

pub struct Stats { /* private fields */ }

Represents the state of the computation of a confidence interval for a proportion.

Examples

let grades = [40, 59, 73, 44, 82, 44, 58, 74, 94, 79, 40, 52, 100, 57, 76, 93, 68, 96, 92, 98, 58, 64, 76, 40, 89, 65, 63, 90, 66, 89];
let stats = proportion::Stats::from_iter(grades.iter().map(|&x| x >= 60));
let confidence = Confidence::new_two_sided(0.95);
let pass_rate_ci = stats.ci(confidence)?;
println!("Pass rate: {}", pass_rate_ci);
assert_abs_diff_eq!(pass_rate_ci, Interval::new(0.4878, 0.8077)?, epsilon = 1e-3);

Panics

• if the number of successes is larger than the population size

Implementations

impl Stats

pub const fn new(population: usize, successes: usize) -> Self

Creates a new statistics object with initial values for the population size and the number of successes. The number of successes must not be larger than the population size.

Complexity: \( O(1) \)

Panics

• if the number of successes is larger than the population size

pub fn population(&self) -> usize

Returns the population size (total number of samples).

Complexity: \( O(1) \)

pub fn successes(&self) -> usize

Returns the number of successes (number of true values found in the sample).

Complexity: \( O(1) \)

pub fn add_success(&mut self)

Add a success to the statistics and updates the population accordingly.

Complexity: \( O(1) \)

pub fn add_failure(&mut self)

Add a failure to the statistics and updates the population accordingly.

Complexity: \( O(1) \)

pub fn is_significant(&self) -> bool

Tests if the conditions for the validity of the Wilson score interval are met. The conditions for the validity of the Wilson score interval are stated as follows: https://www.itl.nist.gov/div898/handbook/prc/section2/prc24.htm

1. The sample size is large enough to ensure that the sampling distribution of the sample proportion is approximately normal (N > 30)
2. The number of successes and failures are large enough to ensure that the sampling distribution of the sample proportion is approximately normal (x > 5 and n - x > 5)

pub fn ci(&self, confidence: Confidence) -> CIResult<Interval<f64>>

Computes the confidence interval over the proportion of true values in a given sample.

Complexity: \( O(1) \)

Arguments

• confidence - the confidence level (must be in (0, 1))

Errors

• TooFewSuccesses - if the number of successes is too small to compute a confidence interval
• TooFewFailures - if the number of failures is too small to compute a confidence interval
• InvalidSuccesses - if the number of successes is larger than the population size
• InvalidConfidenceLevel - if the confidence level is not in (0, 1)

Examples

let data = [
   true, false, true, true, false, true, true, false, true, true,
  false, false, false, true, false, true, false, false, true, false
];
let confidence = Confidence::new_two_sided(0.95);
let stats = proportion::Stats::from_iter(data);
let interval = stats.ci(confidence)?;
assert_abs_diff_eq!(interval, Interval::new(0.299, 0.701)?, epsilon = 1e-2);

Notes

The confidence interval is computed using the function ci_wilson (Wilson score interval).

pub fn extend<I: IntoIterator<Item = bool>>(&mut self, data: I)

Extend the data with additional sample data.

Complexity: \( O(n) \) where \( n \) is the number of samples in data.

Arguments

• data - the sample given as a boolean iterator or slice

Examples

let data = [true, false, true, true, false, true, true, false, true, true];
let mut stats = proportion::Stats::default();
stats.extend(data);
assert_eq!(stats, proportion::Stats::new(10, 7));

pub fn extend_if<T, I, F>(&mut self, data: I, is_success: F)where I: IntoIterator<Item = T>, F: Fn(T) -> bool,

Extend the data with additional sample data and a condition that must be satisfied to be counted as a success.

Complexity: \( O(n) \) where \( n \) is the number of samples in data.

Arguments

• data - the sample given as an iterator or slice
• is_success - a function that returns true if a sample value is a success

Examples

let data = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10];
let mut stats = proportion::Stats::default();
stats.extend_if(data.iter(), |&x| x <= 5);
assert_eq!(stats, proportion::Stats::new(10, 5));

Trait Implementations

impl Add<Stats> for Stats

fn add(self, rhs: Self) -> Self::Output

Combines two statistics objects by adding the number of samples and the number of successes.

Complexity: \( O(1) \)

Examples

let stats1 = proportion::Stats::new(100, 50);
let stats2 = proportion::Stats::new(200, 100);
let stats = stats1 + stats2;
assert_eq!(stats, proportion::Stats::new(300, 150));

type Output = Stats

The resulting type after applying the + operator.

impl AddAssign<Stats> for Stats

fn add_assign(&mut self, rhs: Self)

Combines two statistics objects by adding the number of samples and the number of successes.

Complexity: \( O(1) \)

Examples

let mut stats1 = proportion::Stats::new(100, 50);
let stats2 = proportion::Stats::new(200, 100);
stats1 += stats2;
assert_eq!(stats1, proportion::Stats::new(300, 150));

impl Clone for Stats

fn clone(&self) -> Stats

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for Stats

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

Formats the value using the given formatter.

impl Default for Stats

fn default() -> Stats

Returns the “default value” for a type.

impl FromIterator<bool> for Stats

fn from_iter<I>(iter: I) -> Selfwhere I: IntoIterator<Item = bool>,

Creates a new statistics object with initial values from a Boolean iterator counting the number of successes.

Complexity: \( O(n) \) where \( n \) is the number of samples in iter.

Arguments

• iter - a Boolean iterator or slice

impl PartialEq<Stats> for Stats

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

This method tests for self and other values to be equal, and is used by ==.

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

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Copy for Stats

impl Eq for Stats

impl StructuralEq for Stats

impl StructuralPartialEq for Stats