Wolfram Library Archive

Courseware Demos MathSource Technical Notes
All Collections Articles Books Conference Proceedings

Neighbourhoods of Independence for Random Processes via Information Geometry

Khadiga Arwini
Organization: UMIST
Department: Department of Mathematics
C. T. J. Dodson
Organization: Department of Mathematics, University of Toronto
Journal / Anthology

The Mathematica Journal
Year: 2005
Volume: 9
Issue: 4
Page range: 863-877

In this article we consider the Freund family of distributions as a 4-manifold equipped with Fisher information as Riemannian metric and derive the induced Alpha-geometry, that is, the Alpha-Ricci curvature, the Alpha-scalar curvature, and so forth. We show that the Freund manifold has a positive constant 0-scalar curvature, so geometrically it constitutes part of a sphere. We examine special cases as submanifolds and discuss their geometrical structures; via one submanifold we provide examples of neighbourhoods of the independent case for bivariate distributions having identical exponential marginals. Thus, since exponential distributions complement Poisson point processes, we obtain a means to discuss the neighbourhood of independence for random processes. Our approach using Mathematica handles the Alpha-geometry calculations and graphics effectively and could be transferred to other distribution families.

*Mathematics > Geometry > Computational Geometry
*Mathematics > Probability and Statistics