Type Definition rstat::multivariate::normal::DiagonalNormal

type DiagonalNormal = MvNormal<Vector<f64>, Vector<f64>>;

Multivariate Normal distribution with mean \(\bm{\mu}\) and diagonal covariance matrix \(\mathrm{diag}(\sigma_1, \ldots, \sigma_n)\).

Methods

impl DiagonalNormal

pub fn diagonal<M, S>(mu: M, sigma2_diag: S) -> Result<Self, Error> where
    M: Into<Vector<f64>>,
    S: Into<Vector<f64>>, 

Construct an \(n\)-dimensional DiagonalNormal distribution with mean \ (\bm{\mu}\) and diagonal covariance matrix \((\mathrm{diag}(\sigma_1, \ldots, \sigma_n)\).

Constraints

1. All entries in the covariance are positive real.

Examples

let dist = DiagonalNormal::diagonal(
    vec![0.0, 1.0], vec![0.5, 2.0]
);

assert!(dist.is_ok());

pub fn diagonal_unchecked<M, S>(mu: M, sigma2_diag: S) -> Self where
    M: Into<Vector<f64>>,
    S: Into<Vector<f64>>, 

Construct an \(n\)-dimensional DiagonalNormal distribution with mean \ (\bm{\mu}\) and diagonal covariance matrix \((\mathrm{diag}(\sigma_1, \ldots, \sigma_n)\), without checking for correctness.

Examples

let dist = DiagonalNormal::diagonal_unchecked(
    vec![0.0, 1.0], vec![0.5, 2.0]
);

Trait Implementations

impl ContinuousDistribution for DiagonalNormal

fn pdf(&self, x: &Vec<f64>) -> f64

Evaluates the probability density function (PDF) at (x).

fn log_pdf(&self, x: &Sample<Self>) -> f64

Evaluates the log PDF at (x).

impl Distribution for DiagonalNormal

type Support = ProductSpace<Reals>

Support of sample elements.

type Params = DiagonalNormalParams

Parameter set uniquely defining the instance.

fn support(&self) -> ProductSpace<Reals>

Returns an instance of the support Space, Self::Support.

fn params(&self) -> Self::Params

Returns an instance of the distribution parameters, Self::Params.

fn cdf(&self, _: &Vec<f64>) -> Probability

Evaluates the cumulative distribution function (CDF) at (x).

fn sample<R: Rng + ?Sized>(&self, rng: &mut R) -> Vec<f64>

Draw a random value from the distribution support.

fn into_support(self) -> Self::Support

Converts self into an instance of Self::Support.

fn into_params(self) -> Self::Params

Converts self into an instance of Self::Params.

fn ccdf(&self, x: &Sample<Self>) -> Probability

Evaluates the complementary CDF at (x).

fn log_cdf(&self, x: &Sample<Self>) -> f64

Evaluates the log CDF at (x), i.e. (\ln{F(x)}).

fn log_ccdf(&self, x: &Sample<Self>) -> f64

Evaluates the log complementary CDF at (x), i.e. (\ln{(1 - F(x))}).

fn sample_n<R: Rng + ?Sized>(&self, rng: &mut R, n: usize) -> Vec<Sample<Self>>

Draw n random value from the distribution support.

fn sample_iter<'a, R: Rng + ?Sized>(
    &'a self,
    rng: &'a mut R
) -> Sampler<'a, Self, R>

Draw an indefinite number of random values from the distribution support.

impl From<Params<ArrayBase<OwnedRepr<f64>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<f64>, Dim<[usize; 1]>>>> for DiagonalNormal

fn from(params: DiagonalNormalParams) -> DiagonalNormal

Performs the conversion.

impl Mahalanobis for DiagonalNormal

fn d_mahalanobis_squared(&self, x: &Vec<f64>) -> f64

Compute the squared Mahalanobis distance of sample (\boldsymbol{x}).

fn d_mahalanobis(&self, x: &Sample<Self>) -> f64

Compute the Mahalanobis distance of sample (\boldsymbol{x}).

impl MultivariateMoments for DiagonalNormal

fn mean(&self) -> Vector<f64>

Computes the vector of expected values of the distribution.

fn covariance(&self) -> Matrix<f64>

Computes the covariance matrix of the distribution.

fn variance(&self) -> Vector<f64>

Computes the vector of variances of the distribution.

fn correlation(&self) -> Matrix<f64>

Computes the correlation matrix of the distribution.