scientific article; zbMATH DE number 25932

From MaRDI portal
Publication:3984264





For a polynomial \(P\) in \(R[z_ 1,\dots,z_ n]\), set \(N(x)=\#\{m\in\mathbb{N}^ n:| P(m)|\leq x\}\). There are classical estimates for \(N(x)\) as \(x\to\infty\) when \(P\) is a positive definite quadratic form. The zeta function \(\zeta_ p(s)=\sum_{\min\mathbb{N}^ n,P(m)\neq 0}P(m)^{-s}\) has an analytic continuation to the entire \(s\)- plane and a functional equation and classical Tauberian theory gives \[ N(x)=x^ \rho A(\log x)+O_ \varepsilon(x^{\rho-\mu+\varepsilon}) \] for any \(\varepsilon>0\) with \(\rho,\mu>0\) and \(A\) a non-zero polynomial. This paper and the earlier paper of the author [Compos. Math. 65, 81-120 (1988; Zbl 0637.32002)] show that the same analysis can be carried through when \(P\) is hypoelliptic (that is, \(\lim_{\| x\|\to\infty}DP(x)/p(x)=0\) for any differential monomial \(D)\) and the zeta function includes a positive weight function (that is, \(\sum\varphi(m)P(m)^{-s})\). The generalisation is obtained with the aid of techniques from the theory of several complex variables and elliptic partial differential operators.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3984264)