Series Converg. Hmwk: Determine Convergence of \sum(-1)^n n/(n^p + (-1)^n)

STEMucator
Homework Helper
Messages
2,076
Reaction score
140

Homework Statement



Really tough series to work with.

Determine the convergence ( absolute or conditional ) or divergence of :

##\sum_{n=2}^{∞} \frac{(-1)^n n}{n^p + (-1)^n}##

Homework Equations



?? Series tests?

The Attempt at a Solution



This series is really ugly. I'm not sure how to apply any of the tests I've been given to this particular series.

The ratio test gives me nada, as do Raabe's and Gauss's tests. The n'th root test fails as well as the integral test. The Cauchy Condensation test also yields no results. It's not a geometric series nor can I compare it to anything as is.

The only tests I think which I have left which will be useful is the Alternating series test and the limit form of the comparison test.

I also thought of something which doesn't really involve too much testing. I was thinking I could perhaps separate the positive and negative terms of the series into two other series.

Let :

##\sum u_n = \sum_{n=1}^{∞} \frac{2n}{2^p n^p + 1}##

and

##\sum v_n = \sum_{n=1}^{∞} \frac{2n+1}{(2n+1)^p - 1}##

Then if I show that u_n and v_n both diverge, it would mean that ##\sum |a_n|## diverged and thus ##\sum a_n## would be conditionally convergent.

I'm not sure if I'm approaching this correctly, or if there's something I may not be seeing, but if anyone knows what direction to take it would be appreciated if you could share.
 
Physics news on Phys.org
I would break it into cases according to the range of p. E.g. if p > 0 then for n > N, (np+(-1)n)-1 can be bounded to be within some range of n-p.
 
haruspex said:
I would break it into cases according to the range of p. E.g. if p > 0 then for n > N, (np+(-1)n)-1 can be bounded to be within some range of n-p.

Hey harup, thanks for the input. Just a note that p is never zero.

I was thinking of something and I would like to see if maybe this will help in solving this.

So we have : ##\sum_{n=2}^{∞} \frac{(-1)^n n}{n^p + (-1)^n} = \sum (-1)^na_n##

Now : ## \sum (-1)^na_n ≤ \sum_{n=2}^{∞} \frac{(-1)^n n}{n^p - 1} = \sum (-1)^n b_n##

I'm not sure this will help though.

I don't quite understand how to break it into cases like you're suggesting ( For p > 0 and p < 0 ).
 
Something like this...
(np+(-1)n)-1 = n-p(1+(-1)nn-p)-1
|(np+(-1)n)-1 - n-p| < n-2p+n-3p+... < n-2pα for n > (1-1/α)-1/p
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top