Non-vanishing of high derivatives of Dirichlet \(L\)-functions at the central point (Q1970622): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Removed claims |
||
Property / author | |||
Property / author: Philippe Michel / rank | |||
Property / author | |||
Property / author: Jeffrey M. Vanderkam / rank | |||
Revision as of 10:43, 15 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Non-vanishing of high derivatives of Dirichlet \(L\)-functions at the central point |
scientific article |
Statements
Non-vanishing of high derivatives of Dirichlet \(L\)-functions at the central point (English)
0 references
17 December 2002
0 references
Let \(\chi\) run over all even Dirichlet characters to modulus \(q\), and let \(\Lambda(s,\chi)\) be the `completed' \(L\)-function. It is shown that if \(q\) is large and \(k\) is a fixed positive integer, then a proportion at least \(c_k\) of the derivatives \(\Lambda^{(k)} (\frac 12,\chi)\) are nonzero, with constants \(c_k= \frac 23- O(k^{-2})\). The authors say that a similar result can be proved for odd characters. One can think of this as being analogous to \textit{B. Conrey}'s result [J. Number Theory 16, 49-74 (1983; Zbl 0502.10022)] for the Riemann \(\xi\)-function, which states that a proportion \(1- O(k^{-2})\) of the zeros of \(\xi^{(k)}(s)\) lie on the critical line. It is however disappointing that the constants \(c_k\) only tend to \(\frac 23\) in the present case. The proof depends on a comparison of the first and second moments of \(\Lambda^{(k)} (\frac 12,\chi) M_k (\frac 12,\chi)\), where \(M(s,\chi)\) is a mollifier, taken to have the form \[ M^*(s,\chi)+ (-1)^k \frac{\overline{\tau(\chi)}} {\sqrt{q}} M^*(1-s, \overline{\chi}) \] for a suitable Dirichlet polynomial \(M^*(s,\chi)\).
0 references
Dirichlet \(L\)-functions
0 references
higher derivatives
0 references
nonvanishing
0 references
central point
0 references
mollifier
0 references