Enum QuantileMethod 

pub enum QuantileMethod {
    Linear,
    Lower,
    Higher,
    Nearest,
    Midpoint,
    InvertedCdf,
    AveragedInvertedCdf,
    ClosestObservation,
    InterpolatedInvertedCdf,
    Hazen,
    Weibull,
    MedianUnbiased,
    NormalUnbiased,
}

Interpolation method for quantile_with_method and its percentile / median friends.

Matches all 13 NumPy quantile methods (#462, #566). The continuous methods use the Hyndman-Fan 1996 (alpha, beta) parameterization: virtual_index = n*q + alpha + q*(1 - alpha - beta) - 1 with linear interpolation between the two bracketing sorted elements. The discrete methods compute an integer index via method-specific rules and return the exact sorted element (no interpolation).

Variants

Linear

NumPy default. Continuous with alpha = beta = 1. Returns lo_val * (1 - frac) + hi_val * frac.

Lower

Pick the element at floor(q * (n - 1)) — the lower of the two bracketing sorted elements.

Higher

Pick the element at ceil(q * (n - 1)) — the upper of the two bracketing sorted elements.

Nearest

Pick the sorted element nearest to q * (n - 1), with ties (frac = 0.5) broken to the even index (matches NumPy’s round-half-to-even convention).

Midpoint

Average of the two bracketing sorted elements: (lo + hi) / 2.

InvertedCdf

Discrete method, Hyndman definition 1. Returns sorted[ceil(n * q) - 1] — a step function with jumps at k / n. NumPy’s 'inverted_cdf'.

AveragedInvertedCdf

Discrete method, Hyndman definition 2. Same as InvertedCdf except that when n * q is an integer the result is the average of the two bracketing sorted elements. NumPy’s 'averaged_inverted_cdf'.

ClosestObservation

Discrete method, Hyndman definition 3. sorted[k] where k = round_half_to_even(n*q - 0.5). NumPy’s 'closest_observation'.

InterpolatedInvertedCdf

Continuous method, Hyndman definition 4. alpha = 0, beta = 1. NumPy’s 'interpolated_inverted_cdf'.

Hazen

Continuous method, Hyndman definition 5. alpha = beta = 0.5. NumPy’s 'hazen'.

Weibull

Continuous method, Hyndman definition 6. alpha = beta = 0. NumPy’s 'weibull'.

MedianUnbiased

Continuous method, Hyndman definition 8. alpha = beta = 1/3. NumPy’s 'median_unbiased'.

NormalUnbiased

Continuous method, Hyndman definition 9. alpha = beta = 3/8. NumPy’s 'normal_unbiased'.

Trait Implementations

impl Clone for QuantileMethod

fn clone(&self) -> QuantileMethod

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for QuantileMethod

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

Formats the value using the given formatter.

impl PartialEq for QuantileMethod

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

Tests for self and other values to be equal, and is used by ==.

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

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

impl Copy for QuantileMethod

impl Eq for QuantileMethod

impl StructuralPartialEq for QuantileMethod