Constriction for sets of probabilities

MaRDI QIDQ6423346zbMATH OpenFDO

Authors Michele Caprio, T. Seidenfeld

Publication date 13 January 2023

Mathematics Subject Classification ID

Probabilistic measure theory (60A10) Foundations of probability theory (60A99) Foundations and philosophical topics in statistics (62A01)

Abstract: Given a set of probability measures $mathcalP$ representing an agent's knowledge on the elements of a sigma-algebra $mathcalF$, we can compute upper and lower bounds for the probability of any event $AinmathcalF$ of interest. A procedure generating a new assessment of beliefs is said to constrict $A$ if the bounds on the probability of $A$ after the procedure are contained in those before the procedure. It is well documented that (generalized) Bayes' updating does not allow for constriction, for all $AinmathcalF$. In this work, we show that constriction can take place with and without evidence being observed, and we characterize these possibilities.

This page was built for publication: Constriction for sets of probabilities

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6423346)