Abstract
This chapter presents the notion of statistical models, a structure associated with a family of probability distributions, which can be given a geometric structure. This chapter deals with statistical models given parametrically. By specifying the parameters of a distribution, we determine a unique element of the family. When the family of distributions can be described smoothly by a set of parameters, this can be considered as a multidimensional surface. We are interested in the study of the properties that do not depend on the choice of model coordinates.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
There are several ways of fitting data to a distribution. For instance, minimizing the Kullback–Leibler divergence or the Hellinger distance are just a couple of ways of doing this.
- 2.
Some authors consider a more general definition for parametric models, without requiring properties (i) and (ii). All examples provided in this book satisfy relations (i) and (ii); we required them here for smoothness reasons.
- 3.
The reason why f takes values in \(\mathcal{F}(\mathcal{X}, \mathbb{R})\) is because we consider f(p ξ ) as a real-valued function of x, \(x \in \mathcal{X}\).
- 4.
This follows the 1940s idea of C. R. Rao and H. Jeffreys to use the Fisher information to define a Riemannian metric.
Bibliography
B.S. Clarke, A.R. Barron, Information theoretic asumptotics of Bayes methods. IEEE Trans. Inform. Theory 36, 453–471 (1990)
J.M. Corcuera, F. Giummolé, A Characterization of Monotone and Regular Divergences (Annals of the Institute of Statistical Mathematics, 1998)
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Calin, O., Udrişte, C. (2014). Statistical Models. In: Geometric Modeling in Probability and Statistics. Springer, Cham. https://doi.org/10.1007/978-3-319-07779-6_1
Download citation
DOI: https://doi.org/10.1007/978-3-319-07779-6_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-07778-9
Online ISBN: 978-3-319-07779-6
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)