The Matsumoto-Yor property on trees for matrix variates of different dimensions

The paper is devoted to an extension of the multivariate Matsumoto-Yor (MY) independence property with respect to a tree with p vertices to the case where random variables corresponding to the vertices of the tree are replaced by random matrices. The converse of the p -variate MY property, which characterizes the product of one gamma and p - 1 generalized inverse Gaussian distributions, is extended to characterize the product of the Wishart and p - 1 matrix generalized inverse Gaussian distributions.

[1]  O. Barndorff-Nielsen,et al.  Exponential transformation models , 1982, Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences.

[2]  A. Koudou.  A Matsumoto–Yor property for Kummer and Wishart random matrices , 2012 .

[3]  Ronald W. Butler.  Generalized Inverse Gaussian Distributions and their Wishart Connections , 1998 .

[4]  Marc Yor,et al.  An analogue of Pitman’s 2M – X theorem for exponential Wiener functionals: Part I: A time-inversion approach , 2000, Nagoya Mathematical Journal.

[5]  V. Seshadri,et al.  More on connections between Wishart and matrix GIG distributions , 2008 .

[6]  G. Letac,et al.  The Lukacs-Olkin-Rubin characterization of Wishart distributions on symmetric cones , 1996 .

[7]  P. Vallois,et al.  Independence properties of the Matsumoto-Yor type , 2012, 1203.0381.

[8]  A. Koudou.  A link between the Matsumoto-Yor property and an independence property on trees , 2006 .

[9]  H. Massam,et al.  The Matsumoto--Yor property and the structure of the Wishart distribution , 2006 .

[10]  R. Muirhead.  Aspects of Multivariate Statistical Theory , 1982, Wiley Series in Probability and Statistics.

[11]  C. Herz.  BESSEL FUNCTIONS OF MATRIX ARGUMENT , 1955 .

[12]  David Heckerman,et al.  A Characterization of the Bivariate Normal-Wishart Distribution , 1995 .

[13]  P. Vallois,et al.  Which distributions have the Matsumoto-Yor property? , 2011 .

[14]  H. Massam,et al.  The Matsumoto-Yor property on trees , 2004 .

[15]  J. Wesołowski.  The Matsumoto–Yor independence property for GIG and Gamma laws, revisited , 2002, Mathematical Proceedings of the Cambridge Philosophical Society.

[16]  Christophe Ley,et al.  Characterizations of GIG laws: A survey , 2014 .

[17]  M. Yor,et al.  Interpretation via Brownian motion of some independence properties between GIG and gamma variables , 2003 .

[18]  O. Barndorff-Nielsen,et al.  Trees with random conductivities and the (reciprocal) inverse Gaussian distribution , 1998, Advances in Applied Probability.

[19]  G. Letac,et al.  An independence property for the product of GIG and gamma laws , 2000 .