I try to study the convergence or divergence of the series \sum_{n=2}^{\infty} \frac{(-1)^{n}

Sallie Banks

Sallie Banks

Answered question

2022-01-12

I try to study the convergence or divergence of the series
n=2(1)nn+(1)n. We can see that this series diverges absolutely, because n=21|n+(1)n| is a rearrangement of n=21n which equals to +. Consequently, we can't say for sure that the rearrangement n=2(1)nn+(1)n of n=2(1)nn converges. That's the part where I am stuck. How exactly can I study the convergence (or not) of the first series?
Should I apply Cauchy's criterion and prove that
sn=k=2n(1)kk+(1)k,nN is a basic sequence?

Answer & Explanation

levurdondishav4

levurdondishav4

Beginner2022-01-13Added 38 answers

Without giving a formal proof of it (or even a formal statement of the theorem), the following is true:
You can do any bounded amount of rearranging of an infinite series without affecting its (conditional) convergence/divergence.
For the series at hand, note that
n=2(1)nn+(1)n=13+12+15+14+17+16+
=1213+1415+1617+.
where the ''bounded'' rearranging is to swap pairs of adjacent terms. The rearranged sequence now meets all the criteria of Leibniz's test, so the series is conditionally convergent. (The rearrangement makes it easy to see that the sum is 1ln2.)
A somewhat more formal definition of ''bounded'' rearrangement is that no term is the rearranged series is ever more than B positions from where it appears in the original series, for some fixed positive number B. (In this example, B=1.) Somewhat more formally, a rearrangement aπ(n) or an, where π:NN is a bijection (aka permutation), is ''bounded'' (by B) if |π(n)n|B for all n.
kalupunangh

kalupunangh

Beginner2022-01-14Added 29 answers

Note that
n=2m(1)nn+(1)n=n=2m(1)nn+(1)nn(1)nn(1)n=n=2m(1)nnn21n=2m1n21,
and nn210 monotonically. Try to supply the details in showing that both series on the RHS are convergent.
alenahelenash

alenahelenash

Expert2022-01-24Added 556 answers

Here is a slightly different approach: For large n, we kind of expect (1)nn+(1)n to behave similarly as (1)nn, whose sum we know how to control. In light of this, we study the difference: n=2N((1)nn+(1)n(1)nn)=n=2N1(n+(1)n)n Now it is not hard to verify that the partial sum on the right-hand side converges absolutely as N. Either the direct comparison or limit comparison will work.

Do you have a similar question?

Recalculate according to your conditions!

New Questions in Analysis

Ask your question.
Get an expert answer.

Let our experts help you. Answer in as fast as 15 minutes.

Didn't find what you were looking for?