noise_moments: Moments of stationary NIG and GAL noise

In ngme2: Linear Latent Non-Gaussian Models with Flexible Distributions

Moments of stationary NIG and GAL noise

Description

Compute the mean, variance, skewness, kurtosis, and excess kurtosis for stationary NIG and GAL noise. The distribution is evaluated through the normal mean-variance mixture

X = \delta + \mu V + \sigma \sqrt{V} Z,

where Z \sim N(0, 1). For the package's stationary noise parameterization, delta = -mu, so the mean is zero.

Usage

noise_moments(noise, ...)

## S3 method for class 'ngme_noise'
noise_moments(noise, ...)

noise_nig_moments(delta = -mu, mu = 0, sigma = 1, nu = 1)

noise_gal_moments(delta = -mu, mu = 0, sigma = 1, nu = 1)

Arguments

noise	An ngme_noise object created by noise_nig() or noise_gal().
...	Currently unused.
delta	Location parameter. Defaults to -mu, matching the stationary ngme noise convention.
mu	Shift/skewness parameter in the mixture mean.
sigma	Positive Gaussian scale parameter.
nu	Positive mixing-distribution shape parameter.

Value

A named numeric vector with entries skewness, kurtosis, excess_kurtosis, variance, sd, central_moment3, and central_moment4.

Examples

noise_nig_moments(mu = 1, sigma = 2, nu = 3)
noise_gal_moments(mu = 1, sigma = 2, nu = 3)
noise_moments(noise_nig(mu = 1, sigma = 2, nu = 3))

ngme2 documentation built on May 20, 2026, 9:10 a.m.