Struct stats_ci::mean::Geometric

pub struct Geometric<F: Float> { /* private fields */ }

Represents the state of the computation of the geometric mean. This is a simple implementation that accumulates information about the samples, such as sum and sum of squares. It is implemented as a wrapper around Arithmetic to compute the arithmetic mean of the logarithms of the samples.

It is best used through the StatisticsOps trait.

Implementations

impl<F: Float> Geometric<F>

pub fn new() -> Self

Create a new empty state

Example

let mut stats = mean::Geometric::new();
stats.append(10.)?;
assert_eq!(stats.sample_count(), 1);
assert_abs_diff_eq!(stats.sample_mean(), 10., epsilon = 1e-10);

pub fn append(&mut self, x: F) -> CIResult<()>

Append a new sample to the data

Complexity: \( O(1) \)

pub fn sample_mean(&self) -> F

Geometric mean of the sample

pub fn sample_sem(&self) -> F

Standard error of the geometric mean

Computed as: \( G \frac{s_{\log x_i}}{\sqrt{n-1}} \) where \( G \) is the geometric mean of the sample; \( s_{\log x_i} \) is the estimate of the standard deviation of the logarithms of the samples; and \( n-1 \) is the degree of freedom of the sample data.

Reference

• Nilan Noris. “The standard errors of the geometric and harmonic means and their application to index numbers.” Ann. Math. Statist. 11(4): 445-448 (December, 1940). DOI: 10.1214/aoms/1177731830 JSTOR

pub fn sample_count(&self) -> usize

Number of samples

Complexity: \( O(1) \)

pub fn ci_mean(&self, confidence: Confidence) -> CIResult<Interval<F>>

Confidence interval for the geometric mean

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

Combine two states

Complexity: \( O(1) \)

pub fn ci<I>(confidence: Confidence, data: I) -> CIResult<Interval<F>>where I: IntoIterator<Item = F>,

Compute the confidence interval on the mean of a sample

Arguments

• confidence - The confidence level of the interval
• data - The data to compute the confidence interval on

Output

• Ok(interval) - The confidence interval on the mean of the sample

Errors

• CIError::TooFewSamples - If the input data has too few samples to compute the confidence interval
• CIError::NonPositiveValue - If the input data contains non-positive values when computing harmonic/geometric means.
• CIError::InvalidInputData - If the input data contains invalid values (e.g. NaN)
• CIError::FloatConversionError - If some data cannot be converted to a float

Trait Implementations

impl<F: Float> Add<Geometric<F>> for Geometric<F>

type Output = Geometric<F>

The resulting type after applying the + operator.

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

Performs the + operation.

impl<F: Float> AddAssign<Geometric<F>> for Geometric<F>

fn add_assign(&mut self, rhs: Self)

Performs the += operation.

impl<F: Clone + Float> Clone for Geometric<F>

fn clone(&self) -> Geometric<F>

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<F: Debug + Float> Debug for Geometric<F>

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

Formats the value using the given formatter.

impl<F: Float> Default for Geometric<F>

fn default() -> Self

Returns the “default value” for a type.

impl<F: Float> MeanCI<F> for Geometric<F>

fn ci<I>(confidence: Confidence, data: I) -> CIResult<Interval<F>>where I: IntoIterator<Item = F>,

Compute the confidence interval on the mean of a sample

impl<F: PartialEq + Float> PartialEq<Geometric<F>> for Geometric<F>

fn eq(&self, other: &Geometric<F>) -> 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<F: Float> StatisticsOps<F> for Geometric<F>

fn append(&mut self, x: F) -> CIResult<()>

Append a new sample to the data

fn sample_mean(&self) -> F

Mean of the sample

fn sample_sem(&self) -> F

Standard error of the sample mean

fn ci_mean(&self, confidence: Confidence) -> CIResult<Interval<F>>

Confidence interval of the sample mean

fn sample_count(&self) -> usize

Number of samples

fn ci<I>(confidence: Confidence, data: I) -> CIResult<Interval<F>>where I: IntoIterator<Item = F>,

Compute the confidence interval on the mean of a sample

fn from_iter<I: IntoIterator<Item = F>>(data: I) -> CIResult<Self>

Create a new state and “populates” it with data from an iterator

fn extend<I: IntoIterator<Item = F>>(&mut self, data: I) -> CIResult<()>

Extend the data with additional sample data.

impl<F: Copy + Float> Copy for Geometric<F>

impl<F: Float> StructuralPartialEq for Geometric<F>