Skip to Main Content

Mathematics of Computation

Published by the American Mathematical Society since 1960 (published as Mathematical Tables and other Aids to Computation 1943-1959), Mathematics of Computation is devoted to research articles of the highest quality in computational mathematics.

ISSN 1088-6842 (online) ISSN 0025-5718 (print)

The 2020 MCQ for Mathematics of Computation is 1.78.

What is MCQ? The Mathematical Citation Quotient (MCQ) measures journal impact by looking at citations over a five-year period. Subscribers to MathSciNet may click through for more detailed information.

 

On Newman polynomials which divide no Littlewood polynomial
HTML articles powered by AMS MathViewer

by Artūras Dubickas and Jonas Jankauskas PDF
Math. Comp. 78 (2009), 327-344 Request permission

Abstract:

Recall that a polynomial $P(x) \in \mathbb {Z}[x]$ with coefficients $0, 1$ and constant term $1$ is called a Newman polynomial, whereas a polynomial with coefficients $-1, 1$ is called a Littlewood polynomial. Is there an algebraic number $\alpha$ which is a root of some Newman polynomial but is not a root of any Littlewood polynomial? In other words (but not equivalently), is there a Newman polynomial which divides no Littlewood polynomial? In this paper, for each Newman polynomial $P$ of degree at most $8,$ we find a Littlewood polynomial divisible by $P$. Moreover, it is shown that every trinomial $1+ux^a+vx^b,$ where $a<b$ are positive integers and $u, v \in \{-1,1\},$ so, in particular, every Newman trinomial $1+x^a+x^b,$ divides some Littlewood polynomial. Nevertheless, we prove that there exist Newman polynomials which divide no Littlewood polynomial, e.g., $x^9+x^6+x^2+x+1.$ This example settles the problem 006:07 posed by the first named author at the 2006 West Coast Number Theory conference. It also shows that the sets of roots of Newman polynomials $V_{\mathcal {N}}$, Littlewood polynomials $V_{\mathcal {L}}$ and $\{-1,0,1\}$ polynomials $V$ are distinct in the sense that between them there are only trivial relations $V_{\mathcal {N}}\subset V$ and $V_{\mathcal {L}}\subset V.$ Moreover, $V \ne V_{\mathcal {L}} \cup V_{\mathcal {N}}.$ The proofs of several main results (after some preparation) are computational.
References
Similar Articles
  • Retrieve articles in Mathematics of Computation with MSC (2000): 11R09, 11Y16, 12D05
  • Retrieve articles in all journals with MSC (2000): 11R09, 11Y16, 12D05
Additional Information
  • Artūras Dubickas
  • Affiliation: Department of Mathematics and Informatics, Vilnius University, Naugarduko 24, Vilnius LT-03225, Lithuania -and- Institute of Mathematics and Informatics, Akademijos 4, Vilnius LT-08663, Lithuania
  • Email: arturas.dubickas@mif.vu.lt
  • Jonas Jankauskas
  • Affiliation: Department of Mathematics and Informatics, Vilnius University, Naugarduko 24, Vilnius LT-03225, Lithuania
  • MR Author ID: 825362
  • ORCID: 0000-0001-9770-7632
  • Email: jonas.jankauskas@gmail.com
  • Received by editor(s): December 10, 2007
  • Received by editor(s) in revised form: January 14, 2008
  • Published electronically: May 16, 2008
  • © Copyright 2008 American Mathematical Society
    The copyright for this article reverts to public domain 28 years after publication.
  • Journal: Math. Comp. 78 (2009), 327-344
  • MSC (2000): Primary 11R09, 11Y16, 12D05
  • DOI: https://doi.org/10.1090/S0025-5718-08-02138-8
  • MathSciNet review: 2448710