It plots the log probability trace of matrix Fisher distribution which should close to the maximum value of the logarithm of matrix Fisher distribution, if samples are correctly generated.
The simulated data. An array with at least 2 3x3 matrices.
An arbitrary 3x3 matrix represents the parameter matrix of this distribution.
For a given parameter matrix Fa, maximum value of the logarithm of matrix Fisher distribution is calculated via the form of singular value decomposition of Fa = U Λ V^T which is tr(Λ). Multiply the last column of U by -1 and replace small eigenvalue, say, λ_3 by -λ_3 if | UV^T| = -1.
A plot which shows log probability trace of matrix Fisher distribution. The values are also returned.
R implementation and documentation: Anamul Sajib<firstname.lastname@example.org>
Habeck M. (2009). Generation of three-dimensional random rotations in fitting and matching problems. Computational Statistics, 24(4):719–731.
1 2 3