11
$\begingroup$

Define $\tau(k)$ as follows :

$$\Delta(x)=x\prod_{n=1}^\infty(1-x^n)^{24}=\sum_{k=1}^\infty\tau(k)x^k.$$

This is equivalent to

$$\prod_{n=1}^\infty(1-x^n)^{24}=\sum_{k=1}^\infty\tau(k)x^{k-1}.$$

Let $a(n) = \tau(n)$ mod $(n-1)$ for $n > 1$.

For example (OEIS A299204),

$a(2) = 0, a(3) = 0, a(4) = 1, a(5) = 2$.

Surprisingly, except $ 4 $ and $ 16 $, there is no integer $n \le 10^7$ such that $a(n) = 1$.

Are the only solutions of the equation $a(n) = 1$ $n=4$ and $n=16$?

P.S.

Let $b(k)$ = number of the solution of the equation $a(n) = k$ for $n$ in $[2,10^7]$.

$ \{b(0), b(1), \cdots \} = \{631, 2, 544, 11, 18, 8, 14, 5, 20, 15, 14, 4, 23, 4, 7, 9, 17, 5, 25, 6, 21, 7, 18, 11, 15, 7, 16, 13, 19, 11, 12, 6, 16, 6, 7, 9, 20, 5, 12, 9, 21, 6, 17, 1, 11, 13, 11, 8, 26, 11, 12, 6, 18, 4, 22, 5, 21, 8, 13, 2, 18, 2, 14, 9, 15, 7, 16, 6, 11, 13, 11, 3, 39, 6, 12, 11, 14, 14, 14, 2, 12, 8, 10, 4, 31, 9, 7, 10, 10, 8, 20, 7, 18, 12, 14, 3, 17, 3, 17, 16, \cdots \}.$

Improve this question
$\endgroup$
8
  • 1
    $\begingroup$ Are the solution of a(n)=1 only 4 and 16? $\endgroup$
    – TOM
    Feb 8, 2018 at 14:08
  • 2
    $\begingroup$ I think this is a fine question, although the motivation is missing, and the wording (grammar) could be improved. I am saying this so that the question is not closed (but perhaps improved by some). $\endgroup$
    – GH from MO
    Feb 8, 2018 at 15:08
  • 2
    $\begingroup$ There is definitely something to explain here. The values of $a(n)$ appear t o be fairly uniformly distributed, but if that were the case then the probablility of having no solutions in $[17,10^7]$ would be about $10^{-5.8}$. $\endgroup$
    – Neil Strickland
    Feb 8, 2018 at 16:21
  • 3
    $\begingroup$ I believe it is open whether $\tau(n) = 1$ for any $n$, but it is known that there are only finitely many $n$ for which $\tau(n) = 1$ by work of Murty, Murty and Shorey from 1987. $\endgroup$
    – Jeremy Rouse
    Feb 8, 2018 at 21:27
  • 2
    $\begingroup$ Don't forget that $\Delta\equiv q+q^9+q^{25}+\dots\pmod2$, so the probability is very much biased in favor of even values. Nonetheless it is a fact that $a(n)=3$ occurs quite often for instance. $\endgroup$
    – Henri Cohen
    Feb 8, 2018 at 22:55

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.