0

Can someone find a function f(n) satisfying these bounds? Can you also prove that it does?

$$ \sum\limits_{k=1}^n \Lambda(k) [1-\text{Frac}(\frac{n}{k})][1-\frac{k}{n}\text{Frac}(\frac{n}{k})]=\frac{1}{2}\sum\limits_{k=1}^n \Lambda(k){}\text{}+O(f(n))$$

and

$$ f(n) = o(n)$$,

where $\displaystyle \text{Frac}(\frac{n}{k})$ is the fractional part of $\displaystyle \frac{n}{k}$ and $\Lambda(k)$ is the von Mangoldt function. I would greatly appreciate any help, if someone could even give me an elementary proof I would be willing to do something for them in return.

flag
1 
And the reason you think it is true is? – Anthony Quas Oct 26 at 8:09
2 
Exact duplicate of math.stackexchange.com/questions/221393/… – Greg Martin Oct 26 at 9:31
Question seems answered on math.SE Voting to close as no longer relevant. – quid Nov 20 at 22:57

closed as no longer relevant by quid, Felipe Voloch, Yemon Choi, fedja, Ryan Budney Nov 21 at 3:15

Browse other questions tagged or ask your own question.