1
$\begingroup$

I am looking for the integer solutions for the diophantine equation $y^2 =4n(n-1)(n-2)\cdots (n-k+1)+1$ for a given $k$ where $n>k+1>5$.

In other words, $$y^2=1+4n^{\underline k},\tag{I}$$ where $n^{\underline k} =n(n-1)(n-2)\cdots(n-k+1)$ is the Pochhammer symbol.

Source of the problem:

The main problem is: Solve the equation $$\frac{m!}{(m-2)!}=\frac{n!}{(n-k)!}$$

$m(m-1)$ must be equal to $n^{\underline k}=\frac{n!}{(n-k)!}$

So we must solve the equation

$$m^2-m -n^{\underline k}=0$$ which has integer solution only if $1+4n^{\underline k}$ is a perfect square, where $n^{\underline k} =n(n-1)(n-2)\cdot\ldots\cdot (n-k+1)$ is the Pochhammer symbol.

Known facts:

  1. $n^{\underline k}$ is not a power (see The Product of consecutive integers is never a power by P. Erdos s and J. L. Selfridge), so $4n^{\underline k}$ is not a square.

  2. We know, quadratic-residue $\times$ quadratic-nonresidue = quadratic-nonresidue.

  3. Steven Stadnicki suggested that there is a link between the problem and Singmaster's conjecture, though I don't understand it.

  4. For $k=3$, $E_k (F)$ is an elliptic curve and thus the equation $(I)$ has finite solutions, to be specific $1$. In an elementary way, it can be shown, equation $(I)$ has no solution, since $4P(n,k)+1=(2n^2-6n+2)^2-3$ is a square for finite cases..

  5. See the article of Bugeaud, Y.; Mignotte, M.; Siksek, S.; Stoll, M.; Tengely, Sz.: Integral points on hyperelliptic curves. They succesfully treat the equation $$ n(n+1)=\frac{m!}{60(m-5)!} $$ In fact, it follows that $m\in \lbrace{0, 1, 2, 3, 4, 5, 6, 7, 15, 19\rbrace}$, there is hope for equations of type $\binom{n}{\ell}=\binom{m}{k}$.

  6. For any $k$ there are finitely many solutions by Siegel's theorem (and for $k>4$ there are finitely many rational solutions by Faltings).

Algebraic geometry:

Let us define a curve $E_k (F):=y^2 =4n(n-1)(n-2)\cdots (n-k+1)+1$ for a given $k$ when $F=\mathbb Z$ or $\mathbb Q$. In other words, $E_k (F):=y^2= 1+4n^{\underline k}$.

Question:

  1. Is there infinite integer solutions to the equation $(I)$?
  2. Provided the answer to the question is yes, then, what are the integer solutions of the equation $(I)$?

Please consider non-trivial cases.

If above questions are too difficult to answer please, advise how to approach to this problem and what else are known. Also let me know the computational approach in this problem.

EDIT:

One may find relation to Brocard's problem or $abc$ conjecture, but please note that you are refereeing to the problems which are more general in nature, on the other hand, my problem is more specific.

$\endgroup$
6
  • 1
    $\begingroup$ For any $k$ there are finitely many solutions by Siegel's theorem (and for $k>4$ there are finitely many rational solutions by Faltings). Whether there are infinitely many solutions across all $k$ is most certainly a completely intractable problem, similar to the famous Brocard's problem. I suspect the answer is no, but have no idea how one could prove it. $\endgroup$
    – Wojowu
    Commented Oct 29, 2020 at 11:50
  • $\begingroup$ Also see this post: mathoverflow.net/questions/128036/… $\endgroup$ Commented Oct 29, 2020 at 12:04
  • $\begingroup$ @Wojowu Thank you for your valuable feedback, but I can not see the connection between Brocard's problem and the problem I posted, could you comment on that plz? $\endgroup$ Commented Oct 29, 2020 at 12:08
  • 1
    $\begingroup$ Remove the $4$ and let $k=n$, then you get Brocard. I'm not saying one problem would follow from the other, I'm just saying they look similar and that makes me believe they are similarly hopeless. $\endgroup$
    – Wojowu
    Commented Oct 29, 2020 at 12:14
  • $\begingroup$ For a fixed value of $n-k$, finiteness of solutions follows from the abc conjecture by the result of Luca mentioned here $\endgroup$
    – Wojowu
    Commented Oct 29, 2020 at 12:17

0

You must log in to answer this question.