Would you please give me an example to show that $(P(X), C_R)$ may be a choice structure even if $R$ is not rational (i.e., complete and transitive).


  • For any nonempty set $X$, let $P(X)$ denote the set of all nonempty subsets of X.
  • For any nonempty subset $B$ of $P(X)$, a function $c: B \rightarrow P(X)$ is called a choice function iff $c(A) \subset A$ for all $A \in B$. The pair $(B, c)$ is called a choice structure.
  • For any binary relation $R$ on $X$, define the function $C_R : P(X) \rightarrow P(X) \cup \{\emptyset\}$ as follows: $$C_R (A) = \{x \in A : ( \forall y \in A ) ( xRy ) \}.$$
  • $\begingroup$ I've re-tagged since this doesn't concern the axiom of choice, or proof theory (which, despite the name, is a specific subfield of logic). (And what is the connection with economics?) $\endgroup$ – Noah Schweber Sep 20 '15 at 5:01
  • 1
    $\begingroup$ These concepts are under consumer choice in microeconomics theory. $\endgroup$ – user2521204 Sep 20 '15 at 11:16
  • 1
    $\begingroup$ As it stands, this question has no background, motivation, or context. Such questions are often put on hold until that information is added. $\endgroup$ – Carl Mummert Sep 20 '15 at 20:08

No non-complete relation $R$ can yield a choice structure in this way. Suppose neither $aRb$ nor $bRa$ holds. Then the map $C_R$ is not a choice function: $C_R(\{a, b\})=\emptyset$.

On the other hand, note that $R$ being rational isn't even enough: if $R$ is the usual ordering on $\mathbb{R}$, then $C_R(\mathbb{Z})=\emptyset$.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.