Normal-inverse Gaussian distribution

From Wikipedia, the free encyclopedia

normal-inverse Gaussian (NIG)
Probability density function
Cumulative distribution function
Parameters μ location (real)
α tail heavyness (real)
β asymmetry parameter (real)
δ scale parameter (real)
\gamma = \sqrt{\alpha^2 - \beta^2}
Support x \in (-\infty; +\infty)\!
Probability density function (pdf) \frac{\alpha\delta K_1 \left(\alpha\sqrt{\delta^2 + (x - \mu)^2}\right)}{\pi \sqrt{\delta^2 + (x - \mu)^2}} \; e^{\delta \gamma + \beta (x - \mu)}

Kλ denotes a modified Bessel function of the third kind
Cumulative distribution function (cdf)
Mean μ + δβ / γ
Median
Mode
Variance δα2 / γ3
Skewness 3 \beta /(\alpha \sqrt{\delta \gamma})
Excess kurtosis 3(1 + 4β2 / α2) / (δγ) − 3
Entropy
Moment-generating function (mgf) e^{\mu z + \delta (\gamma - \sqrt{\alpha^2 -(\beta +z)^2})}
Characteristic function

The normal-inverse Gaussian distribution (NIG) is continuous probability distribution that is defined as the normal variance-mean mixture where the mixing density is the inverse Gaussian distribution. The tails of the distribution decrease more slowly than the normal distribution. It is therefore suitable to model phenomena where numerically large values are more probable than is the case for the normal distribution. Examples are returns from financial assets and turbulent wind speeds. The normal-inverse Gaussian distributions form a subclass of the generalised hyperbolic distributions.

The fact that there is a simple expression for the moment generating function implies that simple expressions for all moments are available. The class of normal-inverse Gaussian distributions is closed under convolution in the following sense. If X1 and X2 are independent random variable that are NIG-distributed with the same values of the parameters α and β, but possibly different values of the location and scale parameters, μ1, δ1 and μ2, δ2, respectively, then X1 + X2 is NIG-distributed with parameters α, β,μ1 + μ2 and δ1 + δ2.

The normal-inverse Gaussian distribution can also be seen as the marginal distribution of the normal-inverse Gaussian process which provides an alternative way of explicity constructing it. Starting with a drifting Brownian motion (Wiener process), W(γ)(t) = W(t) + γt, we can define the inverse Gaussian process A_t=\inf\{s>0:W^{(\gamma)}(s)=\delta t\}. Then given a second independent drifting Brownian motion, W^{(\beta)}(t)=\tilde W(t)+\beta t, the normal-inverse Gaussian process is the time-changed process Xt = W(β)(At). The process X(t) at time 1 has the normal-inverse Gaussian distribution described above. The NIG process is a particular instance of the more general class of Lévy processes.

The parameters in the normal-inverse Gaussian distributed is often compiled into a heavyness and skewness plot called the NIG-triangle.

v • d • e
Probability distributions
 
Continuous univariate supported on the whole real line