The Wikipedia entry on the extreme value theorem says that if $f$ is a real-valued continuous function on a closed and bounded interval $[a,b]$, then $f$ must attain a maximum value, i.e. there exists an $x \in [a,b]$ such that $f(x) \geq f(y)$ for all $y \in [a,b]$.

I think that there is a more general version of the extreme value theorem which states a similar result for a closed and bounded subset of $\mathbb{R}^n$. (At least I think I remember hearing about this in a class on metric spaces.) Is there a statement of this more general version of the theorem, hopefully with a reference as well?

  • $\begingroup$ The general thing is compactness. A continuous function on a [non-empty] compact space attains its maximum (and minimum). That's because a continuous image of a compact space is compact. $\endgroup$ – Daniel Fischer Jul 29 '14 at 14:04

Yes, in fact they both come from two standard results in analysis:

Theorem 1: A subset of $\mathbb{R}^n$ is compact if and only if it is closed and bounded.

Theorem 2: Continuous image of a compact set is a compact set.

Corollary: If $f:\mathbb{R}^n\to \mathbb{R}$ is continuous and $A\subset \mathbb{R}^n$ is closed and bounded. By Theorem 1, $A$ is compact. Hence $f(A)$ is a compact subset of $\mathbb{R}$. So $f(A)$ is closed and bounded. That proves the Extreme Value Theorem.


There is this document that states the theorem even more générally:

Theorem 3.4. (The Extreme Value Theorem) If $f : X \to R$ is real valued [edit: and continuous] function from a compact space to the real numbers, then f attains a greatest value, that is there is an $x \in X$ such that $f(x) \geq f(y)$ for all $y \in X$.

  • 1
    $\begingroup$ In proof of Theorem 3.4 the Lemma 3.1 is used (but not referenced with name). In the lemma 3.1 f is correctly required to be continuous. Thus is should be included in Theorem 3.4 that f is required to be continuous. $\endgroup$ – dioid Jul 29 '14 at 14:35
  • $\begingroup$ Thanks, dioid, I've edited my post $\endgroup$ – Igor Jul 29 '14 at 19:59

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.