On P-numbers of quadratic forms

In this paper we introduce P-numbers of quadratic forms over R and study their properties.

Gespeichert in:
Bibliographische Detailangaben
Datum:2009
Hauptverfasser: Bondarenko, V.M., Pereguda, Yu.M.
Format: Artikel
Sprache:English
Veröffentlicht: Інститут математики НАН України 2009
Schlagworte:
Online Zugang:http://dspace.nbuv.gov.ua/handle/123456789/6325
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Zitieren:On P-numbers of quadratic forms / V.M. Bondarenko, Yu.M. Pereguda // Збірник праць Інституту математики НАН України. — 2009. — Т. 6, № 2. — С. 474-477. — Бібліогр.: 1 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id irk-123456789-6325
record_format dspace
spelling irk-123456789-63252010-02-24T12:01:11Z On P-numbers of quadratic forms Bondarenko, V.M. Pereguda, Yu.M. Геометрія, топологія та їх застосування In this paper we introduce P-numbers of quadratic forms over R and study their properties. 2009 Article On P-numbers of quadratic forms / V.M. Bondarenko, Yu.M. Pereguda // Збірник праць Інституту математики НАН України. — 2009. — Т. 6, № 2. — С. 474-477. — Бібліогр.: 1 назв. — англ. 1815-2910 http://dspace.nbuv.gov.ua/handle/123456789/6325 512.5/512.6 en Інститут математики НАН України
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
language English
topic Геометрія, топологія та їх застосування
Геометрія, топологія та їх застосування
spellingShingle Геометрія, топологія та їх застосування
Геометрія, топологія та їх застосування
Bondarenko, V.M.
Pereguda, Yu.M.
On P-numbers of quadratic forms
description In this paper we introduce P-numbers of quadratic forms over R and study their properties.
format Article
author Bondarenko, V.M.
Pereguda, Yu.M.
author_facet Bondarenko, V.M.
Pereguda, Yu.M.
author_sort Bondarenko, V.M.
title On P-numbers of quadratic forms
title_short On P-numbers of quadratic forms
title_full On P-numbers of quadratic forms
title_fullStr On P-numbers of quadratic forms
title_full_unstemmed On P-numbers of quadratic forms
title_sort on p-numbers of quadratic forms
publisher Інститут математики НАН України
publishDate 2009
topic_facet Геометрія, топологія та їх застосування
url http://dspace.nbuv.gov.ua/handle/123456789/6325
citation_txt On P-numbers of quadratic forms / V.M. Bondarenko, Yu.M. Pereguda // Збірник праць Інституту математики НАН України. — 2009. — Т. 6, № 2. — С. 474-477. — Бібліогр.: 1 назв. — англ.
work_keys_str_mv AT bondarenkovm onpnumbersofquadraticforms
AT peregudayum onpnumbersofquadraticforms
first_indexed 2025-07-02T09:15:16Z
last_indexed 2025-07-02T09:15:16Z
_version_ 1836526041834717184
fulltext Çáiðíèê ïðàöüIí-òó ìàòåìàòèêè ÍÀÍ Óêðà¨íè2009, ò.6, �2, 474-477UDC 512.5/512.6V. M. BondarenkoInstitute of Mathemati s, NAS, KyivE-mail: vit-bond�imath.kiev.uaYu. M. PeregudaKorolyov military Institute of national aviation University,ZhytomyrOn PPP -numbers of quadrati forms In this paper we introduce P -numbers of quadratic forms over R and study their properties. In this paper, by a quadratic form we mean a quadratic form over the field of real numbers R f(z) = f(z1, . . . , zn) = n∑ i=1 fiz 2 i + ∑ i<j fijzizj . The set of all such form is denoted by R, and the set of all f(z) ∈ R with f1, . . . , fn = 1 is denoted by R0. Let f(z) ∈ R0 and s ∈ {1, . . . , n}. We introduce the notion of the s-deformation of f(z) as follows: f (s)(z, a) = f (s)(z1, . . . , zn, a) = az2 s + ∑ i6=s z2 i + ∑ i<j fijzizj , where a is a parameter. Denote by F (s) + the set of all b ∈ R such that the form f (s)(z, b) is positive definite, and put F (s) − = R \ F (s) + . In other words, b ∈ F (s) − iff there exists a nonzero vector r = (r1, . . . , rn) ∈ Rn c© V. M. Bondarenko, Yu. M. Pereguda, 2009 On PPP -numbers of quadratic forms 475 such that f (s)(r1, . . . , rn, b) ≤ 0. Further, put m (s) f = sup F (s) − ∈ R ∪∞ (since x ∈ F (s) − implies y ∈ F (s) − for any y < x, this supremum is a limit point). We call m (s) f the s-th P -number of f(z). Proposition 1. Let f(z1, . . . , zn) ∈ R0. Then 1) m (s) f ≥ 0; 2) m (s) f = ∞ if the form f−s(z1, . . . , zs−1, zs+1, . . . , zn) = f(z1, . . . , zs−1, 0, zs+1, . . . , zn) is not positive definite. Both these assertions follow easily from the definitions. Theorem 1. Let f(z1, . . . , zn) ∈ R0 and let m (s) f 6= ∞. Then 1) m (s) f ∈ F (s) − , and consequently m (s) f is the greatest number of F (s) − . 2) the form f (s)(z,m (s) f ) is non-negative definite; Proof. 1) We may assume, without loss of generality, that s = n. Consider the matrix S(a) of the quadratic form f (n)(z, a): S(a) = 1 2   2 f12 . . . f1,n−1 f1n f12 2 . . . f2,n−1 f2n . . . . . . . . . . . . . . . f1,n−1 f2,n−1 . . . 2 fn−1,n f1n f2n . . . fn−1,n 2a   . Demote by ∆k, k = 1, . . . , n− 1, the principal k× k minor of S(a) and by ∆in the (n− 1) × (n− 1) minor of S(a) which is obtained from S(a) by deleting ith arrow and nth column. The determinant 476 V. M. Bondarenko, Yu. M. Pereguda of S(a) is denoted by ∆(a). Then by the well-known formula, ∆(a) = 1/2[(−1)n+1f1n∆1n + (−1)n+2f2n∆2n + · · · · · · + (−1)2n−1fn−1,n∆n−1,1n] + a∆n−1, whence ∆(a) = a∆n−1 +N (∗) where N = 1/2[(−1)n+1f1n∆1n+(−1)n+2f2n∆2n+ · · ·+(−1)2n−1 fn−1,n∆n−1,1n]. By assertion 2) of Proposition 1 the form f−n(z1, . . . , zn−1) is positive definite (since m (n) f 6= ∞). From Silvestr’s criterion of positive definiteness of quadratic forms it follows that ∆1 > 0, . . . ,∆n−1 > 0 . Further, from this criterion it follows that f(z, a) is positive defi- nite if ∆(a) > 0, and is not positive definite if ∆(a) ≤ 0. Conse- quently (see (∗)) F (n) − = {b ∈ R |∆(b) ≤ 0} = {b ∈ R | b∆n−1 ≤ −N} = {b ∈ R | b ≤ −N/∆n−1}. So m (n) f = −N/∆n−1 ∈ F (n) − , as claimed. 2) The first proof. Suppose that f (s)(z,m (s) f ) is not non- negative definite. Then there is a vector r = (r1, . . . , rn) ∈ Rn such that f (s)(r,m (s) f ) = α < 0. Fix 0 < ε < −α. By continuity of f(z, a), there exist δi > 0 for i = 1, . . . , n and δ > 0 such that |f (s)(r1 + µ1, . . . , rn + µn,m (s) f + µ) − f (s)(r1, . . . , rn,m (s) f )| < ε whenever |µi| < δi for i = 1, . . . , n and |µ| < δ. Put µi = 0 for i = 1, . . . , n and fix 0 < µ0 < δ. Then |f (s)(r1, . . . , rn,m (s) f + µ0) − α| < ε. On PPP -numbers of quadratic forms 477 It follows that f (s)(r1, . . . , rn,m (s) f + µ0) − α < ε, whence f (s)(r1, . . . , rn,m (s) f + µ0) < ε+ α < 0. So m (s) f + µ0 ∈ F (s) − , a contradiction to the definition of m (s) f . The second proof. Let s = n. It follows from the proof of assertion 1) (of this theorem) that δ(m (n) f ) = 0. Since ∆1 > 0, . . . , ∆n−1 > 0, the form f (n)(z,m (n) f ) is non-negative definite (see, for example, [1, P.322]). � References [1] V. V. Voevodin Linear algebra. Moskow: Nauka, 1980, 400p. (in Russian).