0
$\begingroup$

Given a continuous function $f: \mathbb R \to \mathbb R $ can we state that $f$ satisfies the conditions of the extreme value theorem at an interval $A=[x_0, x_0 + c], c \in \mathbb R$ as $x_0 \to + \infty$ (or minus infinity)?

Can we state that $f(A) = [m,M]$ where $m,M$ are the extreme values of $f$ at $A$?

There are cases when $m,M$ are e.g. infinitely large but since $\pm \infty$ is a limit point that seems to generate a contradiction. Or must $f$ be defined as $f: \mathbb R \cup \{ \pm \infty \} \to \mathbb R \cup \{ \pm \infty \}$?

$\endgroup$
2
$\begingroup$

With a continujoius extension to the extended reals, you are fine. However, without actually allowing the points o finfinity, even $f(x)=\frac1x$ does not attain its minimum on $[1,\infty)$. Of course it does for every $[1,1+c]$, but that is nothing special

$\endgroup$
  • $\begingroup$ That was my concern. $\endgroup$ – bolzano May 5 '15 at 10:28
0
$\begingroup$

As far as there is no limiting condition , $f(A)=[m,M]$ by extreme value theorem itself . Where do you want to use $x_0 \rightarrow \infty$? It is not clear from your question.

$\endgroup$

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.