Tuesday, 10 January 2017

calculus - Convergence of suminftyn=1fracan1+nan?




I need to prove the convergence/divergence of the series n=1an1+nan based on the convergence/divergence of the series n=1an. It is given that an>0, nN



If the series n=1an is convergent, then from an1+nan<an and the comparison test, we conclude that n=1an1+nan is convergent. However, if the series n=1an is divergent, I have no idea how to prove the convergence/divergence of n=1an1+nan. It is definitely divergent (just take an=1n), but I have no clue how to prove it.


Answer



As you noted, if an=1n, then n=1an1+nan=n=112n diverges.



On the other hand, say a2k=12 and an=2n for n not a power of two. Then
n=1an1+nan<n=12n1+n2n+k=112+2k;


since both right-hand sums are convergent and every term of the left-hand sum is positive, the left-hand sum also converges.




In both cases, an diverges. So merely knowing that an diverges is insufficient to tell you about the convergence behavior of an1+nan.



On the other hand, if an is decreasing and an is divergent, then an1+nan>an1+sn>an2sn for n sufficiently large, where sn=nk=1ak. Moreover, I claim that if an diverges, then so does ansn. Hence, if you add the condition that an is decreasing to the original problem, it will in fact be true that an converges if and only if an1+nan does.



To establish the claim, fix some positive integer N and look at the tail of ansn beginning at the Nth term. For any positive integer k, we have
ki=1aN+isN+i>ki=1aN+isN+k=sN+ksNsN+k=1sNsN+k.


Since the sequence sn is divergent, we can choose k such that sN+k>2sN, and thus 1sNsN+k>12. But N was arbitrary; hence the tails of ansn don't decrease to zero, and so the sum does not converge.


No comments:

Post a Comment

real analysis - How to find limhrightarrow0fracsin(ha)h

How to find limh0sin(ha)h without lhopital rule? I know when I use lhopital I easy get $$ \lim_{h\rightarrow 0}...