Abstract
There is a formula that yields the invariant measure of a form-invariant model \(p(x|\xi )\) in a straightforward way without analysis of the symmetry group, in particular without knowledge of the multiplication function. This is useful because the analysis of the symmetry group may be difficult. This is even of basic importance, because the formula allows one to generalise the definition of the prior \(\mu \) to cases where form invariance does not exist or is not known to exist. Thus we do not require form invariance for the application of Bayes’ theorem.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
R.E. Kass, The Riemannian Structure of Model Spaces: A Geometrical Approach to Inference. Ph.D. thesis, University of Chicago, 1980. See especially pp. 94–95
R.E. Kass, The geometry of asymptotic inference. Stat. Sci. 4, 188–219 (1989)
S.I. Amari, Differential Geometrical Methods in Statistics, Lecture Notes in Statistics, vol. 28 (Springer, Heidelberg, 1985)
C. Radakrishna Rao, Differential metrics in probability spaces based on entropy and divergence measures. Technical Report, Report No. AD-A160301, AFOSR-TR-85-0864, Air Force Office of Scientific Research, Bolling AFB, DC (1985)
R.D. Levine, Geometry in classical statistical thermodynamics. J. Chem. Phys. 84, 910–916 (1986)
C. Radakrishna Rao, Differential metrics in probability spaces, in Shun ichi Amari [12], pp. 218–240
C.C. Rodriguez, Objective Bayesianism and geometry, in Fougère [10], pp. 31–39
C. Villegas, Bayesian inference in models with Euclidean structures. J. Am. Stat. Assoc. 85(412), 1159–1164 (1990)
C.C. Rodriguez, Are we cruising a hypothesis space? in von der Linden et al. [11], pp. 131–139
P.F. Fougère (ed.), Maximum Entropy and Bayesian Methods, Dartmouth, 1989 (Dordrecht, Kluwer,1990)
W. von der Linden, V. Dose, R. Fischer, R. Preuss (eds.), Maximum Entropy and Bayesian Methods, Garching, 1998 (Dordrecht, Kluwer, 1999)
S.I. Amari (ed.), Differential Geometry in Statistical Inference. Lecture Notes and Monograph Series, vol. 10 (Institute of Mathematical Statistics, Hayward, California, 1987)
R.A. Fisher, Theory of statistical information. Proc. Camb. Philos. Soc. 22, 700–725 (1925)
H. Jeffreys, Theory of Probability (Oxford University Press, Oxford, 1939) (2nd ed. 1948; 3rd ed. 1961, here Jeffreys’ rule is found in III §3.10)
H. Jeffreys, An invariant form of the prior probability in estimation problems. Proc. R. Soc. A 186, 453–461 (1946)
T. Chang, C. Villegas, On a theorem of Stein relating Bayesian and classical inferences in group models. Can. J. Stat. 14(4), 289–296 (1986)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Harney, H.L. (2016). Going Beyond Form Invariance: The Geometric Prior. In: Bayesian Inference. Springer, Cham. https://doi.org/10.1007/978-3-319-41644-1_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-41644-1_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-41642-7
Online ISBN: 978-3-319-41644-1
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)