What is it about?

Under a simple class of utility functions, the Bayes-optimal estimates can be formulated using possibility theory instead of probability theory. This interpretation of possibility theory reduces to the likelihood interpretation for a special case of the utility function and the prior possibility measure.

Featured Image

Why is it important?

Applications include not only a formalization of Bayesian model checking and updating but also a method of null hypothesis significance testing in the absence of a completely specified space of alternative hypotheses. That method addresses the replication crisis by weighing relative payoffs for not rejecting the null hypothesis when it holds and or rejecting it when a specified alternative hypothesis holds.

Perspectives

For further reading, follow the "Related papers" link.

David R. Bickel
University of North Carolina at Greensboro

Read the Original

This page is a summary of: Coherent checking and updating of Bayesian models without specifying the model space: A decision-theoretic semantics for possibility theory, International Journal of Approximate Reasoning, March 2022, Elsevier,
DOI: 10.1016/j.ijar.2021.11.006.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page