Struct BayesianGpdEstimator 

pub struct BayesianGpdEstimator { /* private fields */ }

Bayesian GPD estimator for rare-event tail latency modeling.

Uses Peaks Over Threshold (POT) method with Bayesian parameter estimation. Provides credible intervals for extreme quantiles like P99.9, P99.99.

Implementations

impl BayesianGpdEstimator

pub fn new() -> Self

Create a new estimator with default prior.

pub fn with_prior(prior: GpdPrior) -> Self

Create an estimator with custom prior.

pub fn observe(&mut self, values: &[f64], threshold: f64)

Set the threshold and record observations.

Values above threshold are recorded as exceedances.

pub fn posterior_parameters(&self) -> Option<GpdParameters>

Get the posterior mean GPD parameters.

pub fn estimate_quantile( &self, p: f64, credibility: f64, ) -> Option<GpdQuantileEstimate>

Estimate an extreme quantile with credible interval.

Returns (point estimate, lower bound, upper bound) at given credibility level.

pub fn exceedance_probability(&self, value: f64) -> Option<f64>

Estimate the probability of exceeding a given value.

pub fn n_exceedances(&self) -> usize

Get the number of exceedances.

pub fn threshold(&self) -> f64

Get the threshold.

Trait Implementations

impl Clone for BayesianGpdEstimator

fn clone(&self) -> BayesianGpdEstimator

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BayesianGpdEstimator

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

Formats the value using the given formatter.

impl Default for BayesianGpdEstimator

fn default() -> Self

Returns the “default value” for a type.