Abstract
Curved exponential families are so general objects that they seem to have no interesting universal properties. However Abram Kagan [1] discovered in 1985 a remarkable inequality on their Fisher information. This note gives a modern presentation of this result and examples, comparing in particular noncentral and central Wishart distributions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Kagan, A.: An information property of exponential families. Teor. Veroyatnost. i Primenen. 30(4), 783–786 (1985), Theory Probab. Appl. 30(4), 831–835 (1986)
Letac, G., Massam, H.: The noncentral Wishart as an exponential family and its moments. J. Multivariate Anal. 99, 1393–1417 (2008)
Letac, G., Massam, H.: The Laplace transform \((\det s)^{-p}\exp \rm trace\, ( s^{-1})\) and the existence of the non-central Wishart distributions. J. Multivariate Anal. 163, 96–110 (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Letac, G. (2021). The Fisher Information of Curved Exponential Families and the Elegant Kagan Inequality. In: Nielsen, F., Barbaresco, F. (eds) Geometric Science of Information. GSI 2021. Lecture Notes in Computer Science(), vol 12829. Springer, Cham. https://doi.org/10.1007/978-3-030-80209-7_36
Download citation
DOI: https://doi.org/10.1007/978-3-030-80209-7_36
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-80208-0
Online ISBN: 978-3-030-80209-7
eBook Packages: Computer ScienceComputer Science (R0)