A Representation Theorem and Applications to Measure Selection and Noninformative Priors

Manfred Jaeger

A Representation Theorem and Applications to Measure Selection and Noninformative Priors

Manfred Jaeger

Institut for Datalogi

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › peer review

3 Citationer (Scopus)

Abstrakt

Udgivelsesdato: MAR

Originalsprog	Engelsk
Tidsskrift	International Journal of Approximate Reasoning
Vol/bind	38
Udgave nummer	3
Sider (fra-til)	217-243
ISSN	0888-613X
Status	Udgivet - 2005

Citationsformater

@article{1ea28f70922b11daa894000ea68e967b,

title = "A Representation Theorem and Applications to Measure Selection and Noninformative Priors",

abstract = "We introduce a set of transformations on the set of all probability distributions over a finite state space, and show that these transformations are the only ones that preserve certain elementary probabilistic relationships. This result provides a new perspective on a variety of probabilistic inference problems in which invariance considerations play a role. Two particular applications we consider in this paper are the development of an equivariance-based approach to the problem of measure selection, and a new justification for Haldane's prior as the distribution that encodes prior ignorance about the parameter of a multinomial distribution.",

author = "Manfred Jaeger",

year = "2005",

language = "English",

volume = "38",

pages = "217--243",

journal = "International Journal of Approximate Reasoning",

issn = "0888-613X",

publisher = "Elsevier",

number = "3",

}

TY - JOUR

T1 - A Representation Theorem and Applications to Measure Selection and Noninformative Priors

AU - Jaeger, Manfred

PY - 2005

Y1 - 2005

N2 - We introduce a set of transformations on the set of all probability distributions over a finite state space, and show that these transformations are the only ones that preserve certain elementary probabilistic relationships. This result provides a new perspective on a variety of probabilistic inference problems in which invariance considerations play a role. Two particular applications we consider in this paper are the development of an equivariance-based approach to the problem of measure selection, and a new justification for Haldane's prior as the distribution that encodes prior ignorance about the parameter of a multinomial distribution.

AB - We introduce a set of transformations on the set of all probability distributions over a finite state space, and show that these transformations are the only ones that preserve certain elementary probabilistic relationships. This result provides a new perspective on a variety of probabilistic inference problems in which invariance considerations play a role. Two particular applications we consider in this paper are the development of an equivariance-based approach to the problem of measure selection, and a new justification for Haldane's prior as the distribution that encodes prior ignorance about the parameter of a multinomial distribution.

M3 - Journal article

VL - 38

SP - 217

EP - 243

JO - International Journal of Approximate Reasoning

JF - International Journal of Approximate Reasoning

SN - 0888-613X

IS - 3

ER -

A Representation Theorem and Applications to Measure Selection and Noninformative Priors

Abstrakt

Fingeraftryk

Citationsformater