Variance-Gamma Distribution

Requires a Wolfram Notebook System
Interact on desktop, mobile and cloud with the free Wolfram Player or other Wolfram Language products.
In this Demonstration we visualize the probability density function of the variance-gamma distribution, which has parameters (location),
(tail),
(asymmetry), and
(scale). These are all real-valued, with the additional constraints
,
. This distribution has "semi-heavy" tails and has appeared in a diverse range of applications, including models of asset returns in financial markets and turbulent wind speeds.
Contributed by: Peter Falloon (April 2009)
Open content licensed under CC BY-NC-SA
Snapshots
Details
The probability density function for the variance-gamma distribution is given by
,
where is the modified Bessel function of the second kind. It has mean
and variance
.
As , the probability density decays exponentially like
. This is intermediate between the behavior of the normal distribution, which decays more rapidly (like
), and the more extreme "fat tail" behavior of power-law distributions. For this reason, it is sometimes referred to as a "semi-heavy tailed" distribution.
Snapshot 1: as increases, the distribution becomes more rounded around its peak value
Snapshot 2: for non-zero , the distribution becomes skewed, that is, asymmetric
Snapshot 3: as increases, the tails drop off more steeply
See the Wikipedia article on Variance-gamma distribution.
Permanent Citation