×

On the geometry of generalized Gaussian distributions. (English) Zbl 1155.94013

Summary: We consider the space of those probability distributions which maximize the \(q\)-Rényi entropy. These distributions have the same parameter space for every \(q\), and in the \(q=1\) case these are the normal distributions. Some methods to endow this parameter space with a Riemannian metric is presented: the second derivative of the \(q\)-Rényi entropy, the Tsallis entropy, and the relative entropy give rise to a Riemannian metric, the Fisher information matrix is a natural Riemannian metric, and there are some geometrically motivated metrics which were studied by Siegel, Calvo and Oller, Lovrić, Min-Oo and Ruh. These metrics are different; therefore, our differential geometrical calculations are based on a new metric with parameters, which covers all the above-mentioned metrics for special values of the parameters, among others. We also compute the geometrical properties of this metric, the equation of the geodesic line with some special solutions, the Riemann and Ricci curvature tensors, and the scalar curvature. Using the correspondence between the volume of the geodesic ball and the scalar curvature we show how the parameter \(q\) modulates the statistical distinguishability of close points. We show that some frequently used metrics in quantum information geometry can be easily recovered from classical metrics.

MSC:

94A17 Measures of information, entropy
53B21 Methods of local Riemannian geometry
PDFBibTeX XMLCite
Full Text: DOI arXiv

References:

[1] Akin, E., (The Geometry of Population Genetics. The Geometry of Population Genetics, Lecture Notes in Biomathematics, vol. 31 (1979), Springer-Verlag) · Zbl 0437.92016
[2] Amari, S., Differential geometry of curved exponential families — Curvature and information loss, Ann. Statist., 10, 357-385 (1982) · Zbl 0507.62026
[3] Amari, S., (Differential-geometrical Methods in Statistics. Differential-geometrical Methods in Statistics, Lecture Notes in Statistics (1985), Springer: Springer Berlin-New York) · Zbl 0559.62001
[4] Amari, S.; Nagaoka, H., Methods of Information Geometry (2000), American Mathematical Society, Oxford University Press
[5] Atkinson, C.; Mitchell, A. F., Rao’s distance measure, Sankhyā Ser. A, 43, 345-365 (1981) · Zbl 0534.62012
[6] Burbea, J., Informative geometry of probability spaces, Expo. Math., 4, 347-378 (1986) · Zbl 0604.62006
[7] Calvo, M.; Oller, J. M., A distance between multivariate normal distributions based in an embedding into the Siegel group, J. Multivariate Anal., 35, 223-242 (1990) · Zbl 0719.62062
[8] Cencov, N. N., (Statistical Decision Rules and Optimal Inference. Statistical Decision Rules and Optimal Inference, Translations of Mathematical Monographs, vol. 53 (1982), American Mathematical Society: American Mathematical Society Providence, RI)
[9] Costa, J.; Hero, A.; Vignat, C., On solutions to multivariate maximum \(\alpha \)-entropy problems, (Rangarajan, A.; Figuerido, M.; Zerubia, J., EMMCVPR 2003, Lisbon, 7-9 July 2003. EMMCVPR 2003, Lisbon, 7-9 July 2003, Lecture Notes in Computer Science, vol. 2683 (2003), Springer-Verlag: Springer-Verlag Berlin), 211-228
[10] Csiszár, I., Information-type measures of difference of probability distributions and indirect observations, Studia Sci. Math. Hungar., 2, 299-318 (1967) · Zbl 0157.25802
[11] Csiszár, I., On topology properties of \(f\)-divergences, Studia Sci. Math. Hungar., 2, 329-339 (1967) · Zbl 0157.25803
[12] Efron, B., Defining the curvature of a statistical problem (with application to second order efficiency) (with discussion), Ann. Stat., 3, 1189-1242 (1975) · Zbl 0321.62013
[13] Fick, E.; Sauermann, G., (The Quantum Statistics of Dynamic Processes. The Quantum Statistics of Dynamic Processes, Springer Series in Solid-State Sciences, vol. 86 (1990), Springer-Verlag: Springer-Verlag Berlin)
[14] Fisher, R. A., Theory of statistical estimation, Proc. Camb. Phil. Soc., 122, 700-725 (1925) · JFM 51.0385.01
[15] Hellinger, E., Neue Begründung der Theorie quadratischer Formen von unendlich vielen Veränderlichen, J. Reine Angew. Math., 36, 210-271 (1909) · JFM 40.0393.01
[16] James, A. T., The variance information manifold and the function on it, (Krishnaiah, P. K., Multivariate Analysis (1973), Academic Press: Academic Press New York), 157-169
[17] Jeffreys, H., An invariant form for the prior probability in estimation problems, Proc. R. Soc. Lon. Ser. A., 1986, 453-461 (1946) · Zbl 0063.03050
[18] Johnson, O.; Vignat, C., Some results concerning maximum Rényi entropy distributions · Zbl 1135.94317
[19] Kapur, J. N., Generalised Cauchy and Students distributions as maximum entropy distributions, Proc. Nat. Acad. Sci. India Sect. A, 58, 235-246 (1988) · Zbl 0669.60026
[20] Kullback, S., Information Theory and Statistics (1959), John Wiley and Sons: John Wiley and Sons New York · Zbl 0149.37901
[21] Leibler, R. A.; Kullback, S., On information and sufficiency, Ann. Math. Statist., 22, 79-86 (1951) · Zbl 0042.38403
[22] Lutwak, E.; Yang, D.; Zhang, G., Cramer-Rao and moment-entropy inequalities for Rényi entropy and generalized Fisher information, IEEE Trans. Inform. Theory, 51, 473-478 (2005) · Zbl 1205.94059
[23] Mitchell, A. F.S., Statistical manifolds of univariate elliptic distributions, Int. Statist. Rev., 56, 1-16 (1988) · Zbl 0677.62009
[24] Mahalanobis, P. C., On the generalized distance in statistics, Proc. Natl. Inst. Sci. India, 2, 49-55 (1936) · Zbl 0015.03302
[25] Moriguti, S., A lower bound for a probability moment of any absolutely continuous distribution with finite variance, Ann. Math. Statist., 23, 286-289 (1952) · Zbl 0049.36303
[26] Oller, J. M., Information metric for extreme value and logistic probability distributions, Sankhyaā: The Indian Journal of Statistics, Ser. A, 49, 17-23 (1987) · Zbl 0643.62004
[27] Oller, J. M.; Corcuera, J. M., Intrinsic analysis of statistical estimation, Ann. Stat., 23, 1562-1581 (1995) · Zbl 0843.62027
[28] Oller, J. M.; Cuadras, C. M., Rao’s negative multinomial distributions, Sankhyā Ser. A, 47, 75-83 (1985) · Zbl 0595.62044
[29] Petz, D., Geometry of canonical correlation on the state space of a quantum system, J. Math. Phys., 35, 780-795 (1994) · Zbl 0804.53101
[30] Petz, D., Covariance and Fisher information in quantum mechanics, J. Phys. A, 35, 929-939 (2002) · Zbl 1168.81317
[31] Rao, C. R., Information and accuracy attainable in the estimation of statistical parameters, Bull. Calcutta Math. Soc., 37, 81-91 (1945) · Zbl 0063.06420
[32] Rényi, A., On measures of entropy and information, (Neyman, J., Proceedings of the 4th Berkeley Conference on Mathematical Statistics and Probability (1961), University of California Press: University of California Press Berkeley), 547-561
[33] Lovrić, M.; Min-Oo, M.; Ruh, E. A., Multivariate normal distributions parametrized as a Riemannian symmetric space, J. Multivariate Anal., 74, 36-48 (2000) · Zbl 0995.53037
[34] Siegel, C. L., Symplectic Geometry (1964), Academic Press: Academic Press New York · Zbl 0138.31403
[35] Skovgaard, L. T., A Riemannian geometry of the multivariate normal model, Scand. J. Statist., 11, 211-233 (1984) · Zbl 0579.62033
[36] Tsallis, C., Possible generalization of Boltzmann-Gibbs statistics, J. Statist. Phys., 52, 479-487 (1988) · Zbl 1082.82501
[37] Tsallis, C.; Mendes, R. S.; Plastino, A. R., The role of constraints within generalized nonextensive statistics, Comm. Math. Phys., 54, 21-32 (1977)
[38] Zografos, K., On maximum entropy characterization of Pearsons type II and VII multivariate distributions, J. Multivariate Anal., 71, 67-75 (1999) · Zbl 0951.62040
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.