2
$\begingroup$

If $f$ is a continous function and $f(x) \rightarrow c > 0$ when $x \rightarrow \infty$, is it true that the function's anti-derivative $F(x) \rightarrow \infty$ when $x \rightarrow \infty$?

Intuitively, I would say that this is true because I believe that there should be an infinitely large area constrained by the graph of $f(x)$ (which is above the positive $x-axis$) and the positive $x$-axis as $x \rightarrow \infty$. However, I cannot prove this. Is my intuition correct and might a more rigorous argument look like?

$\endgroup$
  • 1
    $\begingroup$ A slightly stronger result holds under the conditions of the question. By L'Hospital's Rule $F(x) /x\to c$ as $x\to \infty $ and your desired conclusion follows. $\endgroup$ – Paramanand Singh Feb 18 at 5:23
3
$\begingroup$

Its true and a rigourous argument is just across the road. Pick $M$ such that for $x>M$, we have the bound $$f(x) > c/2>0.$$ It is enough to show that $\lim_{y\to\infty} \int_M^y f = \infty$. And this is true, because...

$\endgroup$
2
$\begingroup$

You know that there is some $X \in \mathbb{R}$ such that for all $x > X$, $f(x) \in [\frac{1}{2}c, \frac{3}{2}c]$, so $f(x) > \frac{c}{2} > 0$, since $f(x)$ comes arbitrarily close to $c$ if $x$ gets big enough. Hence for all $x > X$: $$ \int_0^x f(s) ds = \int_0^X f(s) ds + \int_X^x f(s) ds > \int_0^X f(s) ds + (x-X)\frac{c}{2}. $$ As $x \to \infty$, the second term blows up, so $F(x) \to \infty$ as $x\to \infty$.

$\endgroup$
2
$\begingroup$

Yes, it's true. Your intuition is correct.Here's a formal way to show it.

Because $f$ converges to $c$, there exists $u\in\mathbb R$ such that if $x\geq u$, then $f(x)\geq \frac c 2$. Then $$F(x)=F(u)+\int_u^x f(t)dt \geq F(u) +\int_u^x \frac{c}{2}dt=F(u) +\frac c 2 (x-u)$$ And the latter $\rightarrow+\infty$.

$\endgroup$
1
$\begingroup$

Yes, that is true. One way to see it is using the FTC: you can say that $f(x)>c/2$ for large enough $x$, say $x>a$) (by the definition of limit) and then $$ F(x)=F(a)+\int\limits_a^xF'(x)\,dx=F(a)+\int\limits_a^x f(x)\,dx $$ $$ \ge F(a)+\int\limits_a^x \frac{c}{2}\,dx \ge F(a)+\frac{c}{2}(x-a) $$ so $F$ grows at least linearly at infinity.

$\endgroup$

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.