Theorem 1.1
6CKPLet 0<\beta<1, and let q=q(T) \le T be a prime satisfying
Then, for every fixed
we have
for all sufficiently large T.
In this paper, we obtain extreme values of Dirichlet L-functions on the critical line with a growing prime modulus. Our result yields a larger lower bound for extreme values than that from Soundararajan’s classical resonance method. Consequently, we show that one branch of the dichotomy established by Bondarenko, Darbar, Hagen, Heap, and Seip (2023) can be realized for this family of Dirichlet L-functions. The proof relies on the resonance method together with precise estimates for GCD sums.
Let 0<\beta<1, and let q=q(T) \le T be a prime satisfying
Then, for every fixed
we have
for all sufficiently large T.
Let q be a prime and let \chi be a non-principal Dirichlet character modulo q. Then, for every real t, we have
Since q is prime and \chi is non-principal, \chi is primitive. Therefore, the result follows from [Iwaniec2004book, Theorem 5.23].
Let \chi be a non-principal Dirichlet character modulo q. Let a>1/2 and set z=x+\mathrm{i} y. Suppose that K(z) is holomorphic in the horizontal strip -a\le y \le 0 satisfying the growth condition
as |x| \to \infty. Then, we have, for every real t,
Put G(z):= L(1/2+\mathrm{i}t+z,\chi)L(1/2-\mathrm{i}t+z,\overline{\chi})K(-\mathrm{i} z). Since \chi is non-principal, both L(s,\chi) and L(s,\overline{\chi}) are entire. Hence G(z) is holomorphic in the strip 0\leq \Re z\leq a. Let Y>2|t|+2, and consider the rectangle with vertices \pm\mathrm{i} Y and a\pm\mathrm{i} Y. By Cauchy’s theorem,
We first show that the integrals over the two horizontal sides tend to zero as Y \to \infty. By Lemma 2.1 and the Phragmén–Lindelöf principle, we have, uniformly for 0\leq \sigma\leq a,
Combining this with the growth condition on K, we obtain
as Y \to \infty. Thus, letting Y \to \infty yields that
Writing z=\mathrm{i} u on the left-hand side and z=a+\mathrm{i} u on the right-hand side, respectively, gives
Since a>1/2, both Dirichlet series on the right-hand side of (?) converge absolutely, so we may integrate term by term. Moreover, we have
Substituting this into (?) proves this lemma.
Let q be a prime and let X\geq 2. Then
The proof of this result follows by applying [Rane1980JLMS, Theorem 3] together with some calculations.
Let \mathcal{M} be a set of integers with |\mathcal{M}| =N. As N \to \infty, we have
Moreover, \mathcal{M} may be chosen so that
where P^+(n) denotes the largest prime factor of n.
This is [delaBreteche2019PLMS, Eq. (1.4)].
The text on this page is from arXiv:2610.02709v1 by Zikang Dong, Qiyu Yang and Shengbo Zhao, under the licence CC0 1.0. The corpus made these changes: it read the items out of the LaTeX source, it expanded the macros of the paper, it put each proof after its statement, and it converted the text to HTML. The formulas are the TeX of the paper, and KaTeX renders them. A formula or a reference can look different from the PDF. The page is published under CC BY 4.0.
The record of each item is in analysis/objects/2610/2610.02709.jsonl.