I have this series

$$\sum_{k=1}^{\infty}\frac{1}{k^{3+\cos k}}$$

I understand that if the exponent is fixed (not a function) and greater than 1 the series converges (p-series) but I can see in wolfram that this series diverges clearly (wolfram says that the comparison test shows that the series diverges... but I dont know what series is using on the comparison).

  • $\begingroup$ Just to be clear, did you mean for the summand to be $\dfrac{1}{a^{3+\cos k}}$ instead of $\dfrac{1}{a(3+\cos k)}$? $\endgroup$ – JimmyK4542 Sep 4 '14 at 6:33
  • $\begingroup$ The first. Sry, I did some mistakes. I put a but I was talking about harmonic-type series and not geonetric-type derivations. $\endgroup$ – Masacroso Sep 4 '14 at 6:38
  • $\begingroup$ Now, you don't have the variable $a$ anywhere in the summation. $\endgroup$ – JimmyK4542 Sep 4 '14 at 6:39
  • $\begingroup$ @JimmyK4542, yes, sorry for the first mistakes... No a, just a derivation of Riemann series with function on exponent. Sry first mistakes. If I put a it isnt what I was trying to compare. $\endgroup$ – Masacroso Sep 4 '14 at 6:40
  • $\begingroup$ Is $n$ a constant? Or did you mean $$\sum_{k=1}^\infty \dfrac{1}{k^{3+\cos(k)}}$$ $\endgroup$ – Robert Israel Sep 4 '14 at 6:41

Since $3 + \cos(k) \ge 2$ for all $k$, $\dfrac{1}{k^{3+\cos(k)}} \le \dfrac{1}{k^2}$, so this series converges by the comparison test. You might have confused Wolfram by using $n$ instead of $k$: if $n$ is a constant the series $\sum_k \dfrac{1}{n^{3+\cos(k)}}$ diverges because the terms are bounded below.

  • $\begingroup$ Yes, surely it was some confuse thing that I did. I was seeing various sums on wolfram and I surely "mixed" things of differents ones. Sry. $\endgroup$ – Masacroso Sep 4 '14 at 6:55

This series is convergent. Since $\cos k\ge -1$, $$\frac{1}{k^{3+\cos k}}\le \frac {1}{k^2}.$$ It would be more interesting if it were $\frac{1}{k^{2+\cos k}}$.

  • $\begingroup$ Can you extend on the example with 2 instead of 3? Please? I will appreciate. $\endgroup$ – Masacroso Sep 4 '14 at 6:53
  • 1
    $\begingroup$ Masacroso: You should be aware now that the series of general term $1/k^{2+\cos k}$ is an entirely different story... hence asking about it on the present page is inappropriate. $\endgroup$ – Did Sep 4 '14 at 7:08

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.