No label defined (Q3030897)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4019163
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 4019163 |
Statements
1987
0 references
critical strip
0 references
Riemann-Siegel formula
0 references
approximate functional equation
0 references
error term
0 references
transformation formula for Dirichlet polynomials
0 references
divisor function
0 references
0.905370831489563
0 references
0.9005330204963684
0 references
0.8956471085548401
0 references
0.8948395252227783
0 references
[For the entire collection see Zbl 0621.00032.] The approximate functional equation for \(\zeta^ 2(s)\) is considered. This is \[ (1)\quad \zeta^ 2(s)=\sum_{n\leq x}d(n)n^{-s}+\chi^ 2(s)\sum_{n\leq y}d(n)n^{s-1}+R(s,x), \] where d(n) is the number of divisors of n, \(s=\sigma +it\), \(0\leq \sigma \leq 1\), \(t\geq 10\), \(x\geq 1\), \(y\geq 1\), \(xy=(t/2\pi)^ 2\), \(\chi (s)=\zeta (s)/\zeta (1-s),\) and R(s,x) may be considered as an error term. A classical result of \textit{E. C. Titchmarsh} [Q. J. Math., Oxf. 9, 109-114 (1938; Zbl 0018.39001)] states that \(R(s,x)=O(x^{1/2-\sigma} \log t).\) In general, this bound is essentially best possible, as shown by the author [see p. 105 of the reviewer's book ``The Riemann zeta-function'' (Wiley New York 1985; Zbl 0556.10026)]. This occurs when \(\sigma =1/2\) for certain values of x and t for which x-t/2\(\pi\) is of the order \(t^{1/2}\), and the present paper expounds this phenomenon in a precise form. Recently the author [Q. J. Math., Oxf. II. Ser. 37, 193-209 (1986; Zbl 0589.10042)] and \textit{Y. Motohashi} [Proc. Japan Acad., Ser. A 59, 393-396, 469-472 (1983; Zbl 0541.10032)] obtained several new bounds for R(s,x) in (1) which improve Titchmarsh's bounds in various ranges of x and t. Another new bound for R(s,x), too complicated to be fully reproduced here, is now proved by the author. The proof is based on a smoothed variant of the approximate functional equation for \(\zeta^ 2(s)\) and a transformation formula for Dirichlet polynomials involving the divisor function d(n). This powerful method, introduced by the author, is fully developed in his recent monograph [Lectures on a method in the theory of exponential sums, Lect. Math. Phys., Math. Tata Inst. Fundam. Res. 80 (Springer (1987)].
0 references