A Posteriori Integration of Probabilities. Elementary Theory

Kryazhimskiy A (2016). A Posteriori Integration of Probabilities. Elementary Theory. Theory of Probability & Its Applications 60 (1): 62-87. DOI:10.1137/S0040585X97T987466.

[img]
Preview
Text
kryazhimskiy2016.pdf - Published Version
Available under License Creative Commons Attribution.

Download (303kB) | Preview

Abstract

An approach to a posteriori integration of probability distributions serving as independent a priori models of observed elementary events from a given finite set of elementary events is proposed. A posteriori integration is understood as an improvement of data given by a priori probabilities. The approach is based on the concept of an a posteriori event in the product of probability spaces associated with a priori probabilities. The conditional probability on the product space that is specified by an a posteriori event determines in a natural way the probability on the set of initial elementary events; the latter is recognized as the result of a posteriori integration of a priori models. Conditions under which the integration improves the informativeness of a priori probabilities are established, algebraic properties of integration as a binary operation on the set of probabilities are studied, and the problem of integral convergence of infinite probability sequences is considered.

Item Type: Article
Uncontrolled Keywords: consistent observational methods, max-measure of concentration, max-compatibility, marginal compatibility, max-concentrator, integration convergence, integration concentration
Research Programs: Advanced Systems Analysis (ASA)
Depositing User: Michaela Rossini
Date Deposited: 29 Mar 2016 13:21
Last Modified: 31 Mar 2017 08:15
URI: http://pure.iiasa.ac.at/12337

Actions (login required)

View Item View Item

International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313