MHB Prove that sin(x)+cos(x) is equal or larger than 1

  • Thread starter Thread starter Yankel
  • Start date Start date
Yankel
Messages
390
Reaction score
0
Hello all

I am trying to prove that

\[sin(x)-cos(x)\geq 1\]

For each x in the interval \[[\frac{\pi }{2},\pi ]\]

I tried doing it by contradiction, what I did was:

Assume

\[sin(x)-cos(x)< 1\]

Then I used the power of 2 on each side of the inequality and got:

\[sin^{2}(x)-2sin(x)cos(x)+cos^{2}(x)<1\]which led me to

\[0<2sin(x)cos(x)\]

which is a contradiction since

\[cos(\frac{\pi }{2})=0\]

I am not sure that what I did is correct or complete. Can you please check my proof and give me your opinion on the matter?

Thank you in advance.
 
Mathematics news on Phys.org
What you did upto the last but one step is perfect, but your showing the contradiction is not that correct, as $\cos (\frac{\pi}{2})=0$ is true only for one value, i.e. at $\pi/2$. The correct contradiction comes from the fact that $2\sin x \cos x\le0 \ \forall x\in[\frac{\pi}{2},\pi]$ from the non-positive nature of $\cos $ function.
 
Thank you. One question if I may.

Logically, if I need to show that for all point in the interval this statement is true, and I showed that there exists a point for which it is not, isn't it enough?
 
Yankel said:
Logically, if I need to show that for all point in the interval this statement is true, and I showed that there exists a point for which it is not, isn't it enough?

Your conclusion only shows that the original statement must be true at $x=\frac{\pi}{2}$. What about the remainder of the interval? Does your conclusion tell us anything about that?
 
I see what you mean ! You are right !

I know notice that my proof is wrong !

\[a<b \sim \rightarrow a^{2}<b^{2}\]

My first step is wrong, isn't it?
 
Last edited:
by the way, is there a way of proving it without using contradiction?

If I know that sin(x) in this interval is between 1 to 0, and cos(x) between 0 to -1, can I claim that this is enough to be considered a proof?
 
Your proof is ok, it's your conclusion that needs work. See vidyarth's post (post #2).

As an alternative, write $\sin(x)-\cos(x)$ as $\sqrt2\sin\left(x-\frac{\pi}{4}\right)$ and note the properties of this sine function over the given interval.
 
Yankel said:
I see what you mean ! You are right !

I know notice that my proof is wrong !

\[a<b \sim \rightarrow a^{2}<b^{2}\]

My first step is wrong, isn't it?

Yes, $a<b$ does not imply $a^2<b^2$ in generality. Consider $a=-4$ and $b=2$. I haven't looked at the problem, but if you want to use that line of thinking, you will need to consider the cases when (a) $a, b>0$, (b) $a, b <0$, (c), ...etc. In particular, for the case of (b), we have $a<b \implies a^2>b^2$.
 
Last edited:
Rido12 said:
Yes, $a<b$ does not imply $a^2<b^2$ in generality. Consider $a=-4$ and $b=2$. I haven't looked at the problem, but if you want to use that line of thinking, you will need to consider the cases when (a) $a, b>0$, (b) $a, b <0$, (c), ...etc. In particular, for the case of (b), we have $a<b \implies a^2>b^2$.

Yes; quite right. My apologies for my blunder. :o
 
  • #10
We can use the triangle inequality.
[TIKZ][blue, ultra thick]
\draw (0,0) -- node[above right] {1} (-3,4) -- node
{$\sin x$} (-3, 0) -- node[below] {$-\cos x$} cycle;
[/TIKZ]

The triangle inequality says that the sum of two sides of a triangle is greater than or equal to the third side.
So:
$$\sin x + (-\cos x) \ge 1$$​
 
  • #11
We could use Lagrange Multipliers...consider the objective function:

$$f(x,y)=\sin(x)-\sin(y)$$

Subject to the constraint:

$$g(x,y)=x+y-\frac{\pi}{2}=0$$

This leads to:

$$\cos(x)=\lambda=-\cos(y)$$

$$\cos(x)+\cos(y)=0$$

Using a sum to product identity, and dividing through by $2$, we obtain:

$$\cos\left(\frac{x+y}{2}\right)\cos\left(\frac{x-y}{2}\right)=0$$

On the given domain, this gives us:

$$x\pm y=\pi\implies x=\pi\pm y$$

Putting this into the constraint, we find:

$$\pi\pm y+y=\frac{\pi}{2}\implies y=-\frac{\pi}{4}\implies x=\frac{3\pi}{4}$$

Thus, the objective function at this critical point is:

$$f\left(\frac{3\pi}{4},-\frac{\pi}{4}\right)=\sqrt{2}$$

At the end-points of the given domain, we find:

$$f\left(\frac{\pi}{2},0\right)=1$$

$$f\left(\pi,-\frac{\pi}{2}\right)=1$$

And so we conclude that on the given domain, we have:

$$f_{\min}=1$$

$$f_{\max}=\sqrt{2}$$
 
  • #12
We can also do a classical function analysis for$f(x)=\sin x - \cos x$ on the domain $[\frac\pi 2,\pi]$.
To find the extrema:
$$f'(x)=0 \quad\Rightarrow\quad \cos x + \sin x = 0 \quad\Rightarrow\quad \sin x=-\cos x
\quad\Rightarrow\quad \tan x=-1\quad\Rightarrow\quad x=\frac{3\pi}{4}$$
So we have one extremum at $x=\frac{3\pi}{4}$ with value $\sqrt 2$.
And we have 2 boundary extrema at $x=\frac{\pi}{2}$ respectively $x=\pi$, both with value $1$.
Therefore:
$$1 \le \sin x - \cos x \le \sqrt 2 \quad\text{ when } x\in [\frac\pi 2,\pi]$$
 
Back
Top