2
$\begingroup$

The twin prime constant is defined as $$ \Pi_2 = \prod_{\substack{p\in\mathcal{P}\\p>2}}\frac{p-2}{p-1}\left(1-\frac{1}{p}\right)^{-1}, $$ where $\mathcal{P}$ is the set of primes. I'm interested in the finite setting, where $P_k$ is the product of the $k$ first primes. The corresponding constant is then
$$ \Pi_2(k) = \prod_{\substack{p>2\\p\mid P_k}}\frac{p-2}{p-1}\left(1-\frac{1}{p}\right)^{-1} $$ My question concerns the related sum $$ \frac{1}{m}\sum_{j=1}^{m} \prod_{\substack{p\mid j \\ p\mid P_k\\p>2}}\frac{p-1}{p-2}. $$ When $m=P_k$ this sum is exactly equal to $\Pi_2(k)^{-1}$, but I would like to know how fast this sum converges.

Question Is it possible to prove an estimate for the error $$ \epsilon_k(m)= \frac{1}{m}\sum_{j=1}^{m} \prod_{\substack{p\mid j \\ p\mid P_k\\p>2}}\frac{p-1}{p-2} - \Pi_2(k)^{-1}? $$ The numerics below (for $k=7$) suggest that $\epsilon_k(m)=O(1/m)$, but I haven't been able to show this theoretically.

enter image description here

$\endgroup$
1
  • $\begingroup$ Your second graph suggests a mean for $m\epsilon_{m}$ of around $-e/2$. $\endgroup$ Commented Feb 20, 2022 at 17:36

0

You must log in to answer this question.

Browse other questions tagged .