Sum and product trigonometry

Geometry Level 3

Is it true that for any whole number n n :

k = 1 n ( 2 sin ( π k 2 n + 1 ) ) 2 = k = 1 n ( 2 sin ( π k 2 n + 1 ) ) 2 ? \displaystyle \sum^n_{k=1} \left(2\sin\left(\frac{\pi k}{2n+1}\right)\right)^2=\prod^n_{k=1}\left(2\sin\left(\frac{\pi k}{2n+1}\right)\right)^2?

So, for example: Is ( 2 sin 3 6 ) 2 + ( 2 sin 7 2 ) 2 = ( 2 sin 3 6 ) 2 × ( 2 sin 7 2 ) 2 (2\sin36^\circ)^2+(2\sin72^\circ)^2=(2\sin36^\circ)^2\times(2\sin72^\circ)^2 true?

Yes No

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

Mark Hennings
Jul 17, 2018

Using de Moivre's Theorem, we see that sin ( 2 n + 1 ) x = j = 0 n ( 1 ) j ( 2 n + 1 2 j + 1 ) cos 2 n 2 j x sin 2 j + 1 x = sin x f ( sin 2 x ) \sin(2n+1)x \; = \; \sum_{j=0}^n (-1)^j \binom{2n+1}{2j+1}\cos^{2n-2j}x\, \sin^{2j+1}x \; = \; \sin x f\big(\sin^2x\big) where f ( X ) f(X) is the degree n n polynomial f ( X ) = j = 0 n ( 1 ) j ( 2 n + 1 2 j + 1 ) ( 1 X ) n j X j f(X) \; = \; \sum_{j=0}^n(-1)^j\binom{2n+1}{2j+1}(1-X)^{n-j}X^j Note that the zeros of f ( X ) f(X) are thus sin 2 k π 2 n + 1 \sin^2\tfrac{k\pi}{2n+1} for k = 1 , 2 , . . . , n k = 1,2,...,n .

The coefficient of X n X^n in f ( X ) f(X) is ( 1 ) n j = 0 n ( 2 n + 1 2 j + 1 ) = ( 1 ) n 2 2 n (-1)^n \sum_{j=0}^n \binom{2n+1}{2j+1} \; = \; (-1)^n 2^{2n} while the coefficient of X n 1 X^{n-1} is j = 0 n 1 ( 1 ) j ( 2 n + 1 2 j + 1 ) ( 1 ) n j 1 ( n j ) = 1 2 ( 1 ) n 1 j = 0 n 1 ( 2 n + 1 ) ! ( 2 j + 1 ) ! ( 2 n 2 j 1 ) ! = 1 2 ( 1 ) n 1 ( 2 n + 1 ) j = 0 n 1 ( 2 n 2 j + 1 ) = 1 2 ( 1 ) n 1 ( 2 n + 1 ) 2 2 n 1 = ( 1 ) n 1 ( 2 n + 1 ) 2 2 n 2 \begin{aligned} \sum_{j=0}^{n-1}(-1)^j \binom{2n+1}{2j+1} (-1)^{n-j-1}(n-j) & = \; \tfrac12(-1)^{n-1}\sum_{j=0}^{n-1}\frac{(2n+1)!}{(2j+1)!(2n-2j-1)!} \; = \; \tfrac12(-1)^{n-1}(2n+1)\sum_{j=0}^{n-1}\binom{2n}{2j+1} \\ & = \; \tfrac12(-1)^{n-1}(2n+1)2^{2n-1} \; = \; (-1)^{n-1}(2n+1)2^{2n-2} \end{aligned} Thus we deduce that k = 1 n ( 2 sin k π 2 n + 1 ) 2 = 4 ( 2 n + 1 ) 2 2 n 2 2 2 n = 2 n + 1 \sum_{k=1}^n \left(2\sin\tfrac{k\pi}{2n+1}\right)^2 \; = \; 4\frac{(2n+1)2^{2n-2}}{2^{2n}} \; = \; 2n+1 On the other hand, the coefficient of X 0 X_0 in f ( X ) f(X) is ( 2 n + 1 1 ) = 2 n + 1 \binom{2n+1}{1} = 2n+1 , and hence k = 1 n ( 2 sin k π 2 n + 1 ) 2 = 2 2 n 2 n + 1 2 2 n = 2 n + 1 \prod_{k=1}^n \left(2\sin\tfrac{k\pi}{2n+1}\right)^2 \; = \; 2^{2n} \frac{2n+1}{2^{2n}} \; = \; 2n+1 Thus both expressions are equal to 2 n + 1 2n+1 .

@Pi Han Goh

Thanks. I knew you would have used the same idea from here . You've got an automatic upvote from me. +1

I was working on a Chebyshev Polynomials approach, but I can't seem to push through. Will revisit this some time soon. I'll tag you if I've made any progress.

Pi Han Goh - 2 years, 10 months ago

Log in to reply

The polynomial trick you link to, and the one I have used here are similar, but subtly different. Using the Chebyshev polynomials of the second kind would involve a slight rewrite of what I have done here, but would certainly be possible. They would give you formulae for the sum and product of the cosines squared, and not the sines , though. It might be possible to translate those results...

Mark Hennings - 2 years, 10 months ago

Log in to reply

I've posted my approach. I don't know why I was so fixated on Chebyshev polynomials, when I can easily invoke 2 famous trigonometric identities.

Regarding your last sentence, yes, I believed that's doable, but it's rather lengthy. Root of unity is the key here. I'll skip it for now.

Pi Han Goh - 2 years, 10 months ago

Please, see also my special matrix solution...

David Ingerman - 2 years, 10 months ago
Pi Han Goh
Jul 19, 2018

Proving the equation is identical to proving that 4 k = 1 n sin 2 ( k π 2 n + 1 ) = 4 n k = 1 n sin 2 ( k π 2 n + 1 ) . 4 \sum_{k=1}^n \sin^2 \left( \dfrac {k \pi}{2n+1} \right) = 4^n \prod_{k=1}^n \sin^2 \left (\dfrac{k \pi}{2n+1} \right) .

Using the double angle formula, sin 2 ( A ) = 1 2 ( 1 cos ( 2 A ) ) \sin^2 (A) = \frac12 (1 - \cos(2A) ) , the left hand side can be expressed as

4 k = 1 n [ 1 2 ( 1 cos ( 2 k π 2 n + 1 ) ) ] = 2 [ n k = 1 n cos ( 2 k π 2 n + 1 ) = 1 / 2 ] = 2 n + 1. 4 \sum_{k=1}^n \left [ \dfrac12 \left (1 - \cos\left( \dfrac {2k \pi}{2n+1} \right) \right) \right ] = 2 \left [ n - \underbrace{\sum_{k=1}^n \cos\left( \dfrac {2k \pi}{2n+1} \right)}_{=-1/2} \right ] = 2n+1.

The final summation above follows from the trigonometric identity:

1 + 2 k = 1 n cos ( k θ ) = k = n n e k i θ = e ( n + 1 / 2 ) θ e i ( n 1 / 2 ) θ e i θ / 2 e i θ / 2 = sin [ ( n + 1 2 ) θ ] sin ( θ 2 ) 1 + 2\sum_{k=1}^n \cos (k\theta) = \sum_{k=-n}^n e^{k i \theta} = \dfrac{e^{(n+ 1/2)\theta} - e^{-i(n-1/2)\theta} }{e^{i\theta/2} - e^{-i \theta /2}} = \dfrac{\sin[(n+\frac12)\theta]}{\sin(\frac\theta2)}

In this case, θ = 2 π 2 n + 1 \theta = \frac{2\pi}{2n+1} , so this summation simplifies to 1 2 -\frac12 .

On the other hand, using the complementary formula, sin ( A ) = sin ( π A ) \sin(A) = \sin(\pi - A) , the right hand side can be expressed as

4 n k = 1 2 n sin ( k π 2 n + 1 ) . 4^n \prod_{k=1}^{2n} \sin \left (\dfrac{k \pi}{2n+1} \right).

This product of sines is also another famous trigonometric identity, which simplifies to 2 n + 1 2 2 n \frac{2n+1}{2^{2n}} . This can proven using the identity k = 1 n 1 sin k π n = n 2 n 1 \displaystyle \prod_{k=1}^{n-1}\sin\frac{k\pi}n = \dfrac n{2^{n-1}} , one of its many proofs can be found here .

Thus, the right hand side simplifies to

4 n 2 n + 1 2 2 n = 2 n + 1 4^n \cdot \dfrac{2n+1}{2^{2n}} = 2n+ 1

as well. Hence, the trigonometric identity in question is true.

Please, also see my special matrix solution...

David Ingerman - 2 years, 10 months ago

Log in to reply

Yeah, either I know too little in linear algebra or your solution is too brief for me to understand.

Pi Han Goh - 2 years, 10 months ago

Log in to reply

The main thing is that elementary symmetric functions (sum, product and etc...) of eigenvalues of a matrix are the coefficients of its characteristic polynomial and can be expressed in terms of its minors determinants (trace, determinant and etc...).

David Ingerman - 2 years, 10 months ago
David Ingerman
Jul 13, 2018

Yes, both expressions are equal to 2 n + 1 2n+1 . It seems that the easiest proof is to notice that ( 2 sin ( k π / n ) ) 2 (2\sin(k\pi/n))^2 are the non-zero eigenvalues with multiplicity 2 of the circulant matrix ( 2 1 0 1 1 2 1 0 0 1 2 1 1 0 1 2 ) . \begin{pmatrix}2 & -1 & 0 & \dots & -1 \\ -1 & 2 & -1 & 0 & \dots \\ 0 & -1 & 2 & -1 & \dots \\ \dots & \dots & \dots & \dots & \dots \\ -1 & \dots & 0 & -1 & 2\end{pmatrix}. Therefore, the sum is half the trace 2 n + 1 2n+1 and the product is the square root of ( 2 n + 1 ) × (2n+1)\times the determinant of any principal 2 n × 2 n 2n\times 2n minor, which is 2 n + 1 2n+1 too (the determinants satisfy simple three-term recurrence, which alows the calculation). @Pi Han Goh @Mark Hennings

Got a proof?

Pi Han Goh - 2 years, 11 months ago

Log in to reply

The sum can be rewritten as a geometric sum. Also for partial proof and relevant discussion see: https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsdGRvbWFpbnxkYXZlZGNsdWJ8Z3g6NmYxZWNiYTA2MzBiMzRhOQ

David Ingerman - 2 years, 10 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...