Struct GaussianCopula 

pub struct GaussianCopula { /* private fields */ }

The Gaussian copula with correlation matrix R.

One draw takes d standard normals z by inverse transform, in order, sets y = L z with L Lᵀ = R, and returns u_j = Φ(y_j). Kendall’s tau between dimensions i and j is (2 / π) asin(R_ij); there is no tail dependence.

use prospicio_core::StreamRng;
use prospicio_prob::copula::{Copula, GaussianCopula};

let c = GaussianCopula::new(&[1.0, 0.5, 0.5, 1.0], 2).unwrap();
let mut u = [0.0; 2];
c.sample(&mut StreamRng::new(1, 0), &mut u);
assert!(u.iter().all(|&x| x > 0.0 && x < 1.0));

Implementations

impl GaussianCopula

pub fn new(correlation: &[f64], dim: usize) -> Result<Self>

A Gaussian copula from a d × d correlation matrix, row-major. It must be symmetric with a unit diagonal and positive definite.

pub fn factor(&self) -> &[f64]

The Cholesky factor L of the correlation matrix, row-major.

Trait Implementations

impl Clone for GaussianCopula

fn clone(&self) -> GaussianCopula

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copula for GaussianCopula

fn dim(&self) -> usize

Number of dimensions.

fn sample(&self, rng: &mut StreamRng, u: &mut [f64])

Fills u (length dim) with one draw of uniforms in (0, 1).

impl Debug for GaussianCopula

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

Formats the value using the given formatter.

impl PartialEq for GaussianCopula

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

Equality operator ==.

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

Inequality operator !=.

impl StructuralPartialEq for GaussianCopula