Monday, 23 September 2013

calculus - Evaluating the sum suminftyn=2ln[11/n2]



For convenience the sum is again
n=2ln[11/n2]=n=2ln(n21)(n2)


I first tried solving using a definite integral, since this seems to make telescoping easier to see,
n=2ln(n21)(n2)=n=2ln(n21)ln(n2)=n=2n21n2dxx

But writing out the first few terms doesn't make any cancellation obvious because of the n2 terms.



I also tried futzing around with log rules and got things down to
n=2ln(n21)(n2)=n=2ln((n1)(n+1))ln(n2)=n=2ln(n1)+ln(n+1)ln(n2)



The first few terms of which are
4n=2ln(n21)(n2)=[ln1+ln32ln2]+[ln2+ln42ln3]+[ln3+ln52ln4]+...=ln2+ln52ln4

Which leads to the guess that
4n=2ln(n21)(n2)=ln(N1)+ln(N+2)2ln(N)=ln((N1)(N+2)N2)0


Which means I'm wrong. Should I soldiering on looking for a pattern through more computation, or is there a more expedient/elegant way to evaluate the sum?


Answer



You were on the right track in writing log(11n2)=log(n+1)log(n)+log(n1)log(n)



Proceeding, we can write



Nn=2log(11n2)=Nn=2(log(n+1)log(n))+Nn=2(log(n1)log(n))=log(N+1)log(2)log(N)




Therefore, we find that



n=2log(11n2)=log(2)


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}...