On the Brück conjecture (Q2800095)

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 6569016
Language Label Description Also known as
default for all languages
No label defined
    English
    On the Brück conjecture
    scientific article; zbMATH DE number 6569016

      Statements

      0 references
      14 April 2016
      0 references
      entire function
      0 references
      Nevanlinna theory
      0 references
      Brück conjecture
      0 references
      shared value
      0 references
      hyper-order
      0 references
      complex linear differential equation
      0 references
      On the Brück conjecture (English)
      0 references
      For an entire function \(f\), the order \(\sigma(f)\) and the hyper-order \(\sigma_2(f)\) are defined by NEWLINE\[NEWLINE \sigma(f) = \limsup_{r\to\infty}{\frac{\log{T(r,f)}}{\log{r}}}\,, \quad \sigma_2(f) = \limsup_{r\to\infty}{\frac{\log{\log{T(r,f)}}}{\log{r}}}\,, NEWLINE\]NEWLINE where \(T(r,f)\) denotes the Nevanlinna characteristic of \(f\). Two non-constant entire functions \(f\) and \(g\) share a value \(a \in \mathbb{C}\) CM (counting multiplicities) if \(f-a\) and \(g-a\) have the same zeros with the same multiplicities. Then a conjecture of the reviewer [Result. Math. 30, 21--24 (1996; Zbl 0861.30032)] states that if a non-constant entire function \(f\) shares a value \(a \in \mathbb{C}\) CM with its derivative \(f'\), and if \(\sigma_2(f)<\infty\) is not a positive integer, then \(f-a=c(f'-a)\), where \(c\) is a nonzero constant.NEWLINENEWLINEThe reviewer already proved the conjecture for \(a=0\). It is also true under the additional assumption \(N(r,0,f')=S(r,f)\) (where \(S(r,f)\) is any quantity satisfying \(S(r,f)=o(T(r,f))\) except possibly on a set of \(r\) of finite linear measure, and \(N(r,0,f')\) is the counting function of the zeros of the derivative \(f'\)) without any growth restrictions. He also gave examples where it does not hold if \(\sigma_2(f)=\infty\) or if \(\sigma_2(f)\) is a positive integer. This conjecture attracted the attention of several authors in the past 20 years.NEWLINENEWLINEFor example, for functions of finite order \(\sigma(f)\) the conjecture was proved by \textit{G.G. Gundersen} and \textit{L.-Z. Yang} [J. Math. Anal. Appl. 223, 88--95 (1998; Zbl 0911.30022)] and also by \textit{L.-Z. Yang} [Kodai Math. J. 22, 458--464 (1999; Zbl 1004.30021)] using another method. A further improvement is due to \textit{Z.-X. Chen} and \textit{K.H. Shon} [Taiwanese J. Math. 8, No. 2, 235--244 (2004; Zbl 1062.30032)] who proved the conjecture for functions of hyper-order \(\sigma_2(f)<\frac{1}{2}\).NEWLINENEWLINEIn this paper, the author proves the conjecture for functions of hyper-order \(\sigma_2(f) \leq \frac{1}{2}\) by studying infinite hyper-order solutions of linear differential equations of the form \(f^{(k)}+A(z)f=Q(z)\), where \(k \geq 2\), \(Q \not\equiv 0\) is an entire function of finite order and \(A\) is a transcendental entire function. Such equations have already been investigated by several authors in previous years. Also, the author replaces the shared value \(a\) by a small function \(a\) with respect to \(f\), that is an entire function with \(T(r,a)=S(r,f)\). More precisely, he proves that if \(f\) is a nonconstant entire function of hyper-order \(\sigma_2(f) \leq \frac{1}{2}\), and if \(a_1\) and \(a_2\) are small functions with respect to \(f\), if \(f-a_1\) and \(f^{(k)}-a_2\) share the value \(0\) CM, then \(f^{(k)}(z)-a_2(z)=c(f(z)-a_1(z))\), where \(c\) is a nonzero constant.
      0 references

      Identifiers