2
$\begingroup$

I've been trying to find a closed form of $\displaystyle \sum_k{\tan{(k)}}$ that contains only elementary functions, and I think I may be onto something. But rather than reinvent the wheel, I want to ensure that this isn't already known.

So, I am specifically interested in the sum of the tangent. However, if it's not too "overbroad", I'm wondering: Which trigonometric summations are known to have these closed forms of elementary functions?

My broader question is to ask about all basic trig functions: sine, arcsine, hyperbolic sine, and inverse hyperbolic sine and their related counterparts like cosine, tangent, secant, cosecant, and cotangent. Which, out of all of these functions, have known elementary closed forms?

If these are known, where I can find references on these sums?

As asked by J.E. Pascoe, I am interested in solutions of

$$\displaystyle \sum_{k=a}^{b}{ \tan{(k)} }$$

where $a$ and $b$ are taken to be naturals. However, if someone can find an answer in some wider domain, this is interesting to me as well.

$\endgroup$
6
  • $\begingroup$ If you mean $\Sigma^{\infty}_{k=0} \tan{k},$ I think that sum does not converge. Can you be more explicit about what sums you're talking about? $\endgroup$ Commented Aug 4, 2014 at 20:24
  • $\begingroup$ Sums of sines or cosines of arithmetic sequences have a closed form, as it is easy to express sines and cosines as exponentials. $\endgroup$
    – user44191
    Commented Aug 4, 2014 at 20:30
  • $\begingroup$ @J.E.Pascoe: I'm interested mainly in finite sums over naturals, but elementary closed forms over other domains are, of course, welcome. $\endgroup$
    – Matt Groff
    Commented Aug 4, 2014 at 20:36
  • 2
    $\begingroup$ See mathoverflow.net/questions/41011/… $\endgroup$
    – Dan Piponi
    Commented Aug 4, 2014 at 20:40
  • $\begingroup$ @DanPiponi: Very interesting, although I'm hoping that by not restricting this question to finding $\Delta_x T=\tan{(x)}$, we may have more freedom. For example, Oleg Eroshkin's answer seems to suggest there is no "nice" closed form solution for the difference delta. $\endgroup$
    – Matt Groff
    Commented Aug 4, 2014 at 21:02

2 Answers 2

2
$\begingroup$

From this answer of mine The antidifference of tangent is:

$$\sum_z\tan z=\sum _{k=1}^{\infty } \left(\psi \left(k \pi -\frac{\pi }{2}+1\right)+\psi \left(k \pi -\frac{\pi }{2}\right)-\psi \left(k \pi -\frac{\pi }{2}+1-z\right)-\psi \left(k \pi -\frac{\pi }{2}+z\right)\right)+C$$

A closed form for it symbolically resolves to

$$\sum_z \tan z=iz-\psi _{e^{2 i}}^{(0)}\left(z+\frac{\pi }{2}\right)+C$$

where the function involved is the q-digamma function (which is not elementary) although the usual definition of q-digamma does not converge at this base.

Thus the answer to your question is no, the antidifference of tangent is not elementary.

But you may be interested in talking about the antidifference taken with step $\pi$ which is more natural than step 1 when speaking about trigonometric functions.

In this case,

$$\Delta_\pi^{-1} \tan x = \frac x\pi \tan x + C$$

which is elementary. Similarly, the antidifference with step 1 of $\tan \pi x$ is also elementary:

$$\sum_x\tan \pi x=x \tan \pi x+C$$

The both identities follow from the period rule.

Regarding other elementary functions you can find some tables in Wikipedia.

$\endgroup$
3
$\begingroup$

Here are some formulas from Jolley, Summation of Series:

$$\sum_1^n\sin k\theta=\sin{(n+1)\theta\over2}\sin{n\theta\over2}\csc{\theta\over2}\tag{417}$$

$$\sum_1^n\cos k\theta=\cos{(n+1)\theta\over2}\sin{n\theta\over2}\csc{\theta\over2}\tag{418}$$

$$\sum_1^na^{k-1}\cosh(k-1)\theta={1-a\cosh\theta-a^n\cosh n\theta+a^{n+1}\cosh(n-1)\theta\over1-2a\cosh\theta+a^2}\tag{714}$$

$\endgroup$
1
  • 2
    $\begingroup$ And these may be proved by induction, using the addition formulas. Or (easier if you know the method) with complex exponentials. $\endgroup$ Commented Dec 29, 2014 at 16:02

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .