Neighbourhoods of independence and associated geometry in manifolds of bivariate Gaussian and Freund distributions

• Autorzy: Khadiga Arwini , Christopher Dodson
• Języki publikacji: EN

Abstrakty

EN

We provide explicit information geometric tubular neighbourhoods containing all bivariate distributions sufficiently close to the cases of independent Poisson or Gaussian processes. This is achieved via affine immersions of the 4-manifold of Freund bivariate distributions and of the 5-manifold of bivariate Gaussians. We provide also the α-geometry for both manifolds. The Central Limit Theorem makes our neighbourhoods of independence limiting cases for a wide range of bivariate distributions; the topological character of the results makes them stable under small perturbations, which is important for applications in models of stochastic processes.

Słowa kluczowe

EN

Information geometry; statistical manifold; bivariate distribution; neighbourhoods of independence; exponential distribution; Freund distribution; Gaussian distribution

Kategorie tematyczne

• 60G05: Foundations of stochastic processes
• 53B20: Local Riemannian geometry

• Wydawca: De Gruyter Open
• Czasopismo: Open Mathematics
• Rocznik: 2007
• Tom: 5
• Numer: 1
• Strony: 50-83

Daty

wydano

2007-03-01

online

2007-03-01

Twórcy

autor

Khadiga Arwini

• University of Manchester

autor

Christopher Dodson

• University of Manchester,

Bibliografia

• [1] S.-I. Amari, O.E. Barndorff-Neilsen, R.E. Kass, S.L. Lauritzen and C.R. Rao: Differential Geometrical Methods in Statistics, Springer Lecture Notes in Statistics 28, Springer-Verlag, Berlin, 1985.
• [2] S-I. Amari and H. Nagaoka: Methods of Information Geometry, American Mathematical Society, Oxford University Press, 2000.
• [3] Khadiga Arwini: Differential geometry in neighbourhoods of randomness and independence, PhD thesis, UMIST, 2004.
• [4] Khadiga Arwini and C.T.J. Dodson: “Information geometric neighbourhoods of randomness and geometry of the McKay bivariate gamma 3-manifold”, Sankhya: Indian Journal of Statistics, Vol. 66(2), (2004), pp. 211–231.
• [5] Khadiga Arwini and C.T.J. Dodson: “Neighbourhoods of independence for random processes via information geometry”, Math. J., Vol. 9(4), (2005).
• [6] Khadiga Arwini, L. Del Riego and C.T.J. Dodson: “Universal connection and curvature for statistical manifold geometry”, Houston J. Math., in press, (2006).
• [7] Y. Cai, C.T.J. Dodson, O. Wolkenhauer and A.J. Doig: “Gamma Distribution Analysis of Protein Sequences shows that Amino Acids Self Cluster”, J. Theoretical Biology, Vol. 218(4), (2002), pp. 409–418.
• [8] C.T.J. Dodson: “Spatial statistics and information geometry for parametric statistical models of galaxy clustering”, Int. J. Theor. Phys., Vol. 38(10), (1999), pp. 2585–2597. http://dx.doi.org/10.1023/A:1026609310371
• [9] C.T.J. Dodson: “Geometry for stochastically inhomogeneous spacetimes”, Nonlinear Analysis, Vol. 47, (2001), pp. 2951–2958. http://dx.doi.org/10.1016/S0362-546X(01)00416-3
• [10] C.T.J. Dodson and Hiroshi Matsuzoe: “An affine embedding of the gamma manifold”, Appl. Sci., Vol. 5(1), (2003), pp. 1–6.
• [11] R.J. Freund: “A bivariate extension of the exponential distribution”, J. Am. Stat. Assoc., Vol. 56, (1961), pp. 971–977. http://dx.doi.org/10.2307/2282007
• [12] T.P. Hutchinson and C.D. Lai: Continuous Multivariate Distributions, Emphasising Applications, Rumsby Scientific Publishing, Adelaide 1990.
• [13] S. Kotz, N. Balakrishnan and N. Johnson: Continuous Multivariate Distributions, Volume 1, 2nd ed., John Wiley, New York, 2000.
• [14] S. Leurgans, T.W.-Y. Tsai and J. Crowley: “Freund’s bivariate exponential distribution and censoring”, In: R.A. Johnson (Ed.): Survival Analysis, IMS Lecture Notes, Hayward, California, Institute of Mathematical Statistics, 1982.
• [15] A.F.S. Mitchell: “The information matrix, skewness tensor and α-connections for the general multivariate elliptic distribution”, Ann. Ins. Stat. Math., Vol. 41, (1989), pp. 289–304. http://dx.doi.org/10.1007/BF00049397
• [16] Y. Sato, K. Sugawa and M. Kawaguchi: The geometrical structure of the parameter space of the two-dimensional normal distribution, Division of information engineering, Hokkaido University, Sapporo, Japan, 1977.
• [17] L.T. Skovgaard: “A Riemannian geometry of the multivariate normal model”, Scandinavian J. Stat., Vol. 11, (1984), pp. 211–223.

Identyfikatory

DOI

10.2478/s11533-006-0034-5