2
$\begingroup$

So, I'm trying to prove the following assertion:

$\lim_{x \to 0} \frac{\sin(x)}{x} = 1$


Proof Attempt:

Let $\epsilon > 0$. Then, we have to show that:

$$\exists \delta > 0 : 0 < |x| < \delta \implies |\frac{\sin(x)}{x} - 1| < \epsilon$$

Let us consider the following:

$$\cos(x) < \frac{\sin(x)}{x} < 1$$

if $0 < |x| < \frac{\pi}{2}$. So, we have:

$$0 < 1 - \frac{\sin(x)}{x} < 1 - \cos(x) = 2\sin^2(\frac{x}{2}) \leq 2 |\sin(\frac{x}{2})| \leq |x| < \delta$$

That's, of course, assuming that $0 < |x| < \frac{\pi}{2}$.

If $\epsilon \geq \frac{\pi}{2}$, then let $\delta = \frac{\pi}{2}$. This would ensure that $1-\frac{\sin(x)}{x}$ is still greater than 0.

If $\epsilon < \frac{\pi}{2}$, then let $\delta = \epsilon$. This would still mean that $0 < |x| < \delta < \frac{\pi}{2}$ so the inequality above would still be satisfied.

This proves the given assertion.

Could someone check my proof above and see if it works or not?

$\endgroup$
  • 1
    $\begingroup$ Compare: math.stackexchange.com/a/329473/721644 $\endgroup$ – Invisible Mar 25 at 18:40
  • $\begingroup$ If you solve the inequality, then $\delta$ works. (: $\endgroup$ – Invisible Mar 25 at 18:44
  • 1
    $\begingroup$ Eyy nice, thanks for the link, I'll check out the discussion there as well. $\endgroup$ – Abhi Mar 25 at 18:45
  • 1
    $\begingroup$ If you accept the "squeeze theorem," then letting $x\to 0$ in $$\cos(x)\le \frac{\sin(x)}{x}\le 1$$suffices. $\endgroup$ – Mark Viola Mar 25 at 19:16
  • $\begingroup$ Can't accept that cos it hasn't been brought up in my text at this point, so I have to prove this in a more elementary way. Just to be sure, is there anything in my proof that could be improved upon? $\endgroup$ – Abhi Mar 25 at 19:18
1
$\begingroup$

First change the name of the variable to $\theta,$ so we have to prove $$\lim_{\theta\to 0}\frac{\sin \theta}{\theta}=1.$$ Take a point $(x,y)$ on the unit circle in the first quadrant. The proof is almost the the same for a point in the 4-th quadrant. Then $\theta$ is the arc length from $(x,y)$ to (1,0) and $\sin \theta=y.$ We need bounds on the arc length. For a non-negative integer $n$ divide the interval $[x,1]$ into $2^n$ equal sub-inervals. Note that we deliberately include the case $n=0$ when the interval is left in one piece. For each division point $x_i$ on the X-axis let $P_i=(x_i,y_i), 0 \le i \le 2^n$ be the corresponding point on the unit circle so that $P_0=(x,y)$ and $P_{2^n}=(1,0).$ For $n \ge 0$ let $$S_n=\sum_{j=1}^{2^n}\text {straight-line lengths from $P_{j-1}$ to $P_j$}.$$ The triangle inequality shows that $S_0,S_1, ...$ is an increasing sequence and that $$S_n \le 1-x+y$$. Thus $\lim_{n \to \infty}S_n$ exists. We define the arc length=$\theta=\lim_{n \to \infty}S_n$Then $$\sqrt{(1-x)^2+y^2}=S_0 \le \theta \le 1-x+y$$. $$\sqrt{(1-\sqrt(1-y^2))^2+y^2}\le \theta \le 1-\sqrt(1-y^2)+y$$ The conditions $\theta \to 0$ and $y \to 0$ are equivalent. All that is needed now is to take the reciprocals in our upper and lower bounds for $\theta$, multiply through by $\sin \theta$, which is the same as $y$ to obtain upper and lower bounds on $\frac {\sin \theta}{\theta}$ in terms of $y.$ With the squeeze law for limits and a little algebra, you can finish the proof.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Thank you for the argument. I can't use the squeeze law to prove this because it hasn't yet been proven in my text. This will be excellent reference for when I do learn it. $\endgroup$ – Abhi Mar 26 at 5:08
1
$\begingroup$

Let's prove an inequality using some geometry :

Let $ \theta \in\mathbb{R} \cdot $ Observe the following figure :

enter image description here

Denoting $ \mathscr{A} $ the red region, (The triangle $ OBC $), $ \mathscr{B} $ the green one, (A part from the unit disc that has a circumference of $ \theta $), $ \mathscr{C} $ the ble one, (The triangle $ OCD $). $ \mathscr{S}_{\mathscr{A}} $, $ \mathscr{S}_{\mathscr{B}} $ and $ \mathscr{S}_{\mathscr{C}} $ are their respective areas.

We have : \begin{aligned} \mathscr{S}_{\mathscr{A}}\leq\mathscr{S}_{\mathscr{B}}\leq\mathscr{S}_{\mathscr{C}}\ \ \ \ \ \\ \iff \frac{\sin{\theta}\cos{\theta}}{2}\leq\frac{\theta}{2}\leq\frac{\tan{\theta}}{2}\ \ \ \\ \iff \ \ \ \ \ \ \ \cos{\theta}\leq\frac{\sin{\theta}}{\theta}\leq\frac{1}{\cos{\theta}}\end{aligned}

Meaning, we have $ \lim\limits_{\theta\to 0}{\frac{\sin{\theta}}{\theta}}=1 \cdot $

| cite | improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ Thank you for the proof but I'm not quite allowed to use the squeeze theorem, which is what you utilized. We haven't proved that yet in the text I'm using. $\endgroup$ – Abhi Mar 26 at 5:06

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.