Kent distribution

In directional statistics, the 5-parameter Fisher–Bingham distribution or Kent distribution, named after Ronald Fisher, Christopher Bingham, and John T. Kent, is a probability distribution on the two-dimensional unit sphere ${\displaystyle S^{2}\,}$ in ${\displaystyle \mathbb {R} ^{3}}$ . It is the analogue on the two-dimensional unit sphere of the bivariate normal distribution with an unconstrained covariance matrix. The Kent distribution was proposed by John T. Kent in 1982, and is used in geology as well as bioinformatics.

Three points sets sampled from the Kent distribution. The mean directions are shown with arrows. The ${\displaystyle \kappa \,}$ parameter is highest for the red set.

The probability density function ${\displaystyle f(\mathbf {x} )\,}$ of the Kent distribution is given by:

${\displaystyle f(\mathbf {x} )={\frac {1}{{\textrm {c}}(\kappa ,\beta )}}\exp\{\kappa {\boldsymbol {\gamma }}_{1}^{T}\cdot \mathbf {x} +\beta [({\boldsymbol {\gamma }}_{2}^{T}\cdot \mathbf {x} )^{2}-({\boldsymbol {\gamma }}_{3}^{T}\cdot \mathbf {x} )^{2}]\}}$

where ${\displaystyle \mathbf {x} \,}$ is a three-dimensional unit vector, ${\displaystyle (.)^{T}}$ denotes the transpose of ${\displaystyle (.)}$, and the normalizing constant ${\displaystyle {\textrm {c}}(\kappa ,\beta )\,}$ is:

${\displaystyle c(\kappa ,\beta )=2\pi \sum _{j=0}^{\infty }{\frac {\Gamma (j+{\frac {1}{2}})}{\Gamma (j+1)}}\beta ^{2j}\left({\frac {1}{2}}\kappa \right)^{-2j-{\frac {1}{2}}}I_{2j+{\frac {1}{2}}}(\kappa )}$

Where ${\displaystyle I_{v}(\kappa )}$ is the modified Bessel function and ${\displaystyle \Gamma (.)}$ is the gamma function. Note that ${\displaystyle c(0,0)=4\pi }$ and ${\displaystyle c(\kappa ,0)=4\pi (\kappa ^{-1})\sinh(\kappa )}$, the normalizing constant of the Von Mises–Fisher distribution.

The parameter ${\displaystyle \kappa \,}$ (with ${\displaystyle \kappa >0\,}$ ) determines the concentration or spread of the distribution, while ${\displaystyle \beta \,}$ (with ${\displaystyle 0\leq 2\beta <\kappa }$ ) determines the ellipticity of the contours of equal probability. The higher the ${\displaystyle \kappa \,}$ and ${\displaystyle \beta \,}$ parameters, the more concentrated and elliptical the distribution will be, respectively. Vector ${\displaystyle \gamma _{1}\,}$ is the mean direction, and vectors ${\displaystyle \gamma _{2},\gamma _{3}\,}$ are the major and minor axes. The latter two vectors determine the orientation of the equal probability contours on the sphere, while the first vector determines the common center of the contours. The 3×3 matrix ${\displaystyle (\gamma _{1},\gamma _{2},\gamma _{3})\,}$ must be orthogonal.