I was looking at the following proof of the Extreme Value Theorem.

$\textbf{Question:}$ Why does $f(x)$ is continuous and $g(x)=\frac{1}{M-f(x)} \implies g(x)$ is continuous?

We know $f(x)$ is continuous here which means when $a\in X$

$\forall \epsilon > 0$ $\exists \delta >0$ $\forall x\in X$, $|x-a|<\delta\rightarrow |f(x)-f(a)|<\epsilon$.

My Attempt to Try to Prove $g(x)$ is Continuous: Let $\epsilon >0$. Working backward to find a $\delta$.

We want to show $|g(x)-g(a)|<\epsilon$. In other words, $-\epsilon<g(x)-g(a)<\epsilon$. Continuing, $-\epsilon<\frac{1}{M-f(x)}-\frac{1}{M-f(a)}<\epsilon$. I don't know where to go from here...


Follows immediately from actually simplifying that expression that you have for $g(x)-g(a)$. In the numerator, you will get a term that goes to $0$ due to the continuity of $f$, and the denominator is not infinitesimal (note that they define $M>f$). Then, for a given pair $(\epsilon,\delta)$ for $f$, you will be able to find a corresponding $(\epsilon,\delta')$ for $g$.

  • $\begingroup$ Doing that I get $\frac{1}{M-f(x)}-\frac{1}{M-f(a)}=\frac{M-f(a)-(M-f(x))}{(M-f(x))\cdot (M-f(a))}=\frac{f(x)-f(a)}{(M-f(x))\cdot (M-f(a))}\implies -\epsilon\cdot (M-f(x))\cdot (M-f(a))<f(x)-f(a)<\epsilon \cdot (M-f(x))\cdot (M-f(a))$. Then, would I take $\delta=\epsilon \cdot (M-f(x))\cdot (M-f(a))$? $\endgroup$ – W. G. Dec 3 '19 at 1:37
  • 1
    $\begingroup$ Nevermind this answers my question. Thank you. $\endgroup$ – W. G. Dec 3 '19 at 1:49

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.