2
$\begingroup$

The lemma says:

Lemma: Suppose $f$ is differentiable and $f'$ is increasing. If $a < b$ and $f(a) = f(b)$, then $f(x) < f(a) = f(b)$ for $a < x < b$.

Proof: Suppose first that $f(x) > f(a) = f(b)$ for some $x$ in $(a, b)$. Then the maximum of $f$ on $[a, b]$ occurs at some point $x_0$ in $(a, b)$ with $f(x_0) > f(a)$ and of course $f'(x_0) = 0$.

Then the proof continues:

On the other hand, applying the Mean Value Theorem to the interval $[a, x_0]$, we find that there is $x_1$ with $a < x_1 < x_0$ and

$$f'(x_1) = \frac{f(x_0) - f(a)}{x_0 - a} > 0$$

Contradicting the fact that $f'$ is increasing. [...]

I understand the lemma tells us that if $f'$ is increasing, then the graph of $f$ will be below any horizontal secant line through $(a, f(a))$ and $(b, f(b))$.

I can also see the proof is by contradiction here, we will show that if $f(x) > f(a) = f(b)$ then $f'$ can't be increasing. Also clear is that $f'(x_0) = 0$ since we are looking at a local maximum for the first case of the proof.

What I'm having trouble with is what the application of the Mean Value Theorem. For the first case where we work with the local maximum, I understand we find a point $a < x_1 < x_0$ with the same slope as the line connecting the endpoints of the interval $[a, x_0]$ would have, but how does it show $f'$ can't be increasing and how do we know $f'(x_1) > 0$?

$\endgroup$

1 Answer 1

2
$\begingroup$

First, it shows that $f'$ can't always be increasing because it is positive at $x_1$ but it's $0$ at $x_0 \gt x_1$, so $f'$ must have decreased. As for knowing $f'\left(x_1\right) \gt 0$, since the maximum of $f$ occurs at $x_0$, then $f\left(x_0\right) \gt f\left(a\right)$. Also, by the condition for $x_0$, you have that $x_0 \gt a$. Thus, both the numerator and denominator of $f'\left(x_1\right)$ are positive, making the result $\gt 0$.

$\endgroup$
5
  • $\begingroup$ I think I get it a bit more, although I find the details still hard to grok. But generally, the proof relies on the fact that if $f'$ is increasing, then $f''$ is positive on the whole interval, is that right? $\endgroup$ Commented Jan 1, 2019 at 2:40
  • $\begingroup$ Because if we find a point where $f'$ is positive and then another point after that one where $f' = 0$, then $f'$ must have been decreasing between those two points and then $f''$ would have been negative. $\endgroup$ Commented Jan 1, 2019 at 2:42
  • $\begingroup$ @Max Even though $f'$ is increasing, there is no guarantee that $f''$ even exists, but it must be positive if it does. As for your second comment, you are correct as well re: $f''$, but the main issue is that $f'$ decreasing contradicts the original problem statement, so the original assumption must be incorrect, proving the desired result using proof by contradiction. $\endgroup$ Commented Jan 1, 2019 at 2:45
  • 1
    $\begingroup$ Thanks again, i need to let it sink in, but I feel I can get it now. Cheers! $\endgroup$ Commented Jan 1, 2019 at 2:58
  • $\begingroup$ @Max You are welcome. I'm glad I was able to be of some help. $\endgroup$ Commented Jan 1, 2019 at 2:59

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.