I wish to prove divergence of $$\sum_{n=3}^\infty \frac{\sqrt{n}+2}{n-2}$$

I wish to do so by comparison, since $n\geq 3$: $$\sum_{n=3}^\infty \frac{\sqrt{n}+2}{n-2} > \sum_{n=3}^\infty \frac{1+2}{n-2}>\sum_{n=3}^\infty \frac{3}{n}>\sum_{n=3}^\infty \frac{1}{n} \rightarrow \infty$$ And the harmonic series is divergent, so if we just remove finitely many terms, we still have that it is divergent, because divergence is determined "in the tail". We have a divergent minorant series and hence the original series diverges to $\infty$.

Is this approach fine, or is there some more elegant method, this was about the simplest thing I could think of.

Alternatively we have: $$\sum_{n=3}^\infty \frac{\sqrt{n}+2}{n-2} > \sum_{n=3}^\infty \frac{\sqrt{n}+2}{n}=\sum_{n=3}^\infty \frac{1}{\sqrt{n}}+ \frac{2}{n}\rightarrow \infty$$

  • 3
    $\begingroup$ What's your question? $\endgroup$ – user23793 Nov 29 '18 at 21:52
  • 1
    $\begingroup$ The summand is $O(\frac{1}{\sqrt{n}})$, which diverges $\endgroup$ – Alex Nov 29 '18 at 21:53
  • $\begingroup$ My bad. I had the question in my head but did not actually write it out xD $\endgroup$ – Wesley Strik Nov 29 '18 at 21:55
  • $\begingroup$ Apparently $\frac{1}{\sqrt{n}}$ diverges, I was not sure about this, but with that method I could just carry out the division and conclude this immediately I suppose $\endgroup$ – Wesley Strik Nov 29 '18 at 21:56

Not clear what your question is, but your answer is correct.

Your approach is fine. Comparison test would be the proper test to use.

One can also show that $$\sum _{n=3}^{\infty \:}\frac{\sqrt{n}+2}{n-2}\ge \sum _{n=3}^{\infty \:}\frac{\sqrt{n}+2}{n}$$

and show that the rightmost sum is diverging via the integral test.

  • $\begingroup$ That's a nice alternative, we haven't practised a lot with this yet, but this method would work very well :) $\endgroup$ – Wesley Strik Nov 29 '18 at 22:01
  • $\begingroup$ Right now I am still trying to figure out which method would be best suited for which situation, I still feel I'm just throwing all I know at a question most of the time - but I'm getting better. $\endgroup$ – Wesley Strik Nov 29 '18 at 22:02
  • $\begingroup$ Generally if you can't evaluate it directly, but you can "visualize" the series diverges, then this test is good. $\endgroup$ – K Split X Nov 29 '18 at 22:03
  • $\begingroup$ Intuiton can be developed, you're right ;) "seeing it" and a proof are sometimes closer than it seems. $\endgroup$ – Wesley Strik Nov 29 '18 at 22:05

$$\sum_{n=3}^\infty \frac{\sqrt{n}+2}{n-2}=\sum_{n=3}^\infty \frac1{\sqrt{n}-2}$$ and the terms are of order $n^{-1/2}$.



$$\frac{\sqrt{n}+2}{n-2} \sim \frac {\sqrt n}n=\frac1{\sqrt n}$$

the series diverges by limit comparison test with $\sum \frac1{\sqrt n}$.


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.