Talk:Aumann's agreement theorem

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

What can't they disagree about?[edit]

The text does not make clear what they can't disagree about. Why is: "Chocolate is the best!" "Vanilla is the best!" not a disagreement? Why isn't: "Unicorns exist." "No, they do not." a disagreement? --Moritzgedig (talk) 15:23, 23 March 2022 (UTC)[reply]

Well, technically they are disagreements, and they can be made to disappear if the hypotheses of the theorem are not violated, at least in principle. You can decide on (A) by agreeing to a certain definition of "best" (best for you, best for the majority, best according to some physical parameters, etc.) and on a method to measure it (for example, if you settle on majority definition, interview a sample of people that you both recognize as representative of the wider population). (B) is pretty much the same: what do the parts mean by "exist"? How do you operationalize it? Gnc9400 (talk) 16:32, 20 April 2022 (UTC)[reply]

A *finite* set of states?[edit]

The article says that Aumann's model consists of (among other things) "a finite set of states ." Does really have to be finite? I couldn't find this restriction mentioned in Aumann's original 1976 paper, and moreover, the "main" example in that paper is about the parameter of a coin, which is an infinite set (the interval between 0 and 1). Not being an economist, I may be missing something, but maybe somebody more knowledgeable than me will find it appropriate to delete the word "finite." Huckbuc (talk) Huckbuc (talk) 08:11, 16 October 2022 (UTC)[reply]