2
$\begingroup$

Let E be an elliptic curve over Q with positive rank and trivial torsion structure. Is there any sort of upper bound (conjectural or unconditional) on the regulator of E in terms of the conductor of E? (For lower bound we have Lang's conjecture.)

$\endgroup$
4
  • 1
    $\begingroup$ In any specific case, you can get a conjectural upper bound from the leading coefficient of the $L$-function at $s=1$. Is that what you are after or are you looking for a theoretical uniform bound? If the latter, then dependent on what? It is very unlikely that there is a constant uniform bound that works for all curves. $\endgroup$
    – Alex B.
    Nov 11, 2010 at 5:12
  • $\begingroup$ I was looking for an upper bound in terms of the conductor or the discriminant of E. I've been trying to figure out if I can give some sort of a bound for $\lim_{s\rightarrow 1}L^{(r)}(s)$ using the conductor or the discriminant of the elliptic curve, but I don't know if that's reasonable or not. $\endgroup$
    – Soroosh
    Nov 12, 2010 at 19:05
  • $\begingroup$ There is Lang's conjecture on the regulator times size of Sha. See his book "Survey of Diophantine Geometry", pp. 99, Chapter 3 section 6, conjecture 6.3. If you don't have access to the book (and can't search google), I can post it here. $\endgroup$ Nov 12, 2010 at 20:10
  • 2
    $\begingroup$ $L^r(1)$ can be bounded by complex analysis, in particular, if you don't care about optimal results, use the Phragmen Lindelof theorem. I think it should give $L(1)$ is less than $N^{1/4}$ times some power of $\log N$. Each deriviative introduces another log via a Cauchy derivative formula, among other methods. More explicitly, $L(s)$ is bounded by $(\log N)^{2?}$ on the $s=3/2$ line via a limiting Euler product (or do it on $s=3/2+1/\log N$), and then by the functional equation, you get a bound $\sqrt N$ times this on the $s=1/2$ line, and apply convexity to get $N^{1/4}$ in the middle. $\endgroup$
    – Junkie
    Feb 28, 2011 at 12:30

0

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.