Showing 1–50 of 59 results for author: Kostov, V P

Numerical approach for solving problems arising from polynomial analysis

Authors: Yousra Gati, Vladimir Petrov Kostov, Mohamed Chaouki Tarchi

Abstract: This paper deals with the use of numerical methods based on random root sampling techniques to solve some theoretical problems arising in the analysis of polynomials. These methods are proved to be practical and give solutions where traditional methods might fall short. This paper deals with the use of numerical methods based on random root sampling techniques to solve some theoretical problems arising in the analysis of polynomials. These methods are proved to be practical and give solutions where traditional methods might fall short. △ Less

Submitted 14 April, 2025; originally announced April 2025.

On the location of the complex conjugate zeros of the partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We prove that for any $q\in (0,1)$, all complex conjugate pairs of zeros of the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ with non-negative real part belong to the half-annulus $\{$Re$(x)\geq 0,~1<|x|<5\}$, where the outer radius cannot be replaced by a number smaller than $e^{π/2}=4.810477382\ldots$, and that for $q\in (0,0.2^{1/4}=0.6687403050\ldots ]$, $θ(q,.)$ has no… ▽ More We prove that for any $q\in (0,1)$, all complex conjugate pairs of zeros of the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ with non-negative real part belong to the half-annulus $\{$Re$(x)\geq 0,~1<|x|<5\}$, where the outer radius cannot be replaced by a number smaller than $e^{π/2}=4.810477382\ldots$, and that for $q\in (0,0.2^{1/4}=0.6687403050\ldots ]$, $θ(q,.)$ has no zeros with non-negative real part. The complex conjugate pairs of zeros with negative real part belong to the left open half-disk of radius $49.8$ centered at the origin. △ Less

Submitted 27 January, 2025; originally announced January 2025.

Comments: 18 pages, 1 figure

MSC Class: 26A06

On universal sign patterns

Authors: Vladimir Petrov Kostov

Abstract: We consider polynomials $Q:=\sum _{j=0}^da_jx^j$, $a_j\in \mathbb{R}^*$, with all roots real. When the {\em sign pattern} $σ(Q):=({\rm sgn}(a_d),{\rm sgn}(a_{d-1})$, $\ldots$, ${\rm sgn}(a_0))$ has $\tilde{c}$ sign changes, the polynomial $Q$ has $\tilde{c}$ positive and $d-\tilde{c}$ negative roots. We suppose the moduli of these roots distinct. The {\em order} of these moduli is defined when in… ▽ More We consider polynomials $Q:=\sum _{j=0}^da_jx^j$, $a_j\in \mathbb{R}^*$, with all roots real. When the {\em sign pattern} $σ(Q):=({\rm sgn}(a_d),{\rm sgn}(a_{d-1})$, $\ldots$, ${\rm sgn}(a_0))$ has $\tilde{c}$ sign changes, the polynomial $Q$ has $\tilde{c}$ positive and $d-\tilde{c}$ negative roots. We suppose the moduli of these roots distinct. The {\em order} of these moduli is defined when in their string as points of the positive half-axis one marks the places of the moduli of negative roots. A sign pattern $σ^0$ is {\em universal} when for any possible order of the moduli there exists a polynomial $Q$ with $σ(Q)=σ^0$ and with this order of the moduli of its roots. We show that when the polynomial $P_{m,n}:=(x-1)^m(x+1)^n$ has no vanishing coefficients, the sign pattern $σ(P_{m,n})$ is universal. We also study the question when $P_{m,n}$ can have vanishing coefficients. △ Less

Submitted 29 May, 2024; originally announced May 2024.

On a problem inspired by Descartes' rule of signs

Authors: Vladimir Petrov Kostov

Abstract: We study real univariate polynomials with non-zero coefficients and with all roots real, out of which exactly two positive. The sequence of coefficients of such a polynomial begins with $m$ positive coefficients followed by $n$ negative followed by $q$ positive coefficients. We consider the sequence of moduli of their roots on the positive real half-axis; all moduli are supposed distinct. We mark… ▽ More We study real univariate polynomials with non-zero coefficients and with all roots real, out of which exactly two positive. The sequence of coefficients of such a polynomial begins with $m$ positive coefficients followed by $n$ negative followed by $q$ positive coefficients. We consider the sequence of moduli of their roots on the positive real half-axis; all moduli are supposed distinct. We mark in this sequence the positions of the moduli of the two positive roots. For $m=n=2$, $n=q=2$ and $m=q=2$, we give the exhaustive answer to the question which the positions of the two moduli of positive roots can be. △ Less

Submitted 22 April, 2024; originally announced April 2024.

Journal ref: Modern Mathematical Methods 2(2) 2024, p. 103-116

Degree $6$ hyperbolic polynomials and orders of moduli

Authors: Yousra Gati, Vladimir Petrov Kostov, Mohamed Chaouki Tarchi

Abstract: We consider real univariate degree $d$ real-rooted polynomials with non-vanishing coefficients. Descartes' rule of signs implies that such a polynomial has $\tilde{c}$ positive and $\tilde{p}$ negative roots counted with multiplicity, where $\tilde{c}$ and $\tilde{p}$ are the numbers of sign changes and sign preservations in the sequence of its coefficients, $\tilde{c}+\tilde{p}=d$. For $d=6$, we… ▽ More We consider real univariate degree $d$ real-rooted polynomials with non-vanishing coefficients. Descartes' rule of signs implies that such a polynomial has $\tilde{c}$ positive and $\tilde{p}$ negative roots counted with multiplicity, where $\tilde{c}$ and $\tilde{p}$ are the numbers of sign changes and sign preservations in the sequence of its coefficients, $\tilde{c}+\tilde{p}=d$. For $d=6$, we give the exhaustive answer to the question: When the moduli of all $6$ roots are distinct and arranged on the real positive half-axis, in which positions can the moduli of the negative roots be depending on the signs of the coefficients? △ Less

Submitted 23 October, 2023; originally announced October 2023.

MSC Class: 26C10; 30C15

Journal ref: Math. Commun. 29 (2024), 163-176

On Descartes' rule of signs for hyperbolic polynomials

Authors: Vladimir Petrov Kostov

Abstract: We consider univariate real polynomials with all roots real and with two sign changes in the sequence of their coefficients which are all non-vanishing. One of the changes is between the linear and the constant term. By Descartes' rule of signs, such degree $d$ polynomials have $2$ positive and $d-2$ negative roots. We consider the sequences of the moduli of their roots on the real positive half-a… ▽ More We consider univariate real polynomials with all roots real and with two sign changes in the sequence of their coefficients which are all non-vanishing. One of the changes is between the linear and the constant term. By Descartes' rule of signs, such degree $d$ polynomials have $2$ positive and $d-2$ negative roots. We consider the sequences of the moduli of their roots on the real positive half-axis. When the moduli are distinct, we give the exhaustive answer to the question at which positions can the moduli of the two positive roots be. △ Less

Submitted 5 April, 2023; originally announced April 2023.

Journal ref: Serdica Math. J. 49 (2023) 251-268

Beyond Descartes' rule of signs

Authors: Vladimir Petrov Kostov

Abstract: We consider real univariate polynomials with all roots real. Such a polynomial with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients has $c$ positive and $p$ negative roots counted with multiplicity. Suppose that all moduli of roots are distinct; we consider them as ordered on the positive half-axis. We ask the question: If the positions of the sign changes are known… ▽ More We consider real univariate polynomials with all roots real. Such a polynomial with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients has $c$ positive and $p$ negative roots counted with multiplicity. Suppose that all moduli of roots are distinct; we consider them as ordered on the positive half-axis. We ask the question: If the positions of the sign changes are known, what can the positions of the moduli of negative roots be? We prove several new results which show how far from trivial the answer to this question is. △ Less

Submitted 10 February, 2023; originally announced February 2023.

Journal ref: Constructive Mathematical Analysis, vol. 6 issue 2 (2023) 128-141

Further than Descartes' rule of signs

Authors: Yousra Gati, Vladimir Petrov Kostov, Mohamed Chaouki Tarchi

Abstract: The {\em sign pattern} defined by the real polynomial $Q:=Σ_{j=0}^da_jx^j$, $a_j\neq 0$, is the string $σ(Q):=({\rm sgn(}a_d{\rm )},\ldots ,{\rm sgn(}a_0{\rm )})$. The quantities $pos$ and $neg$ of positive and negative roots of $Q$ satisfy Descartes' rule of signs. A couple $(σ_0,(pos,neg))$, where $σ_0$ is a sign pattern of length $d+1$, is {\em realizable} if there exists a polynomial $Q$ with… ▽ More The {\em sign pattern} defined by the real polynomial $Q:=Σ_{j=0}^da_jx^j$, $a_j\neq 0$, is the string $σ(Q):=({\rm sgn(}a_d{\rm )},\ldots ,{\rm sgn(}a_0{\rm )})$. The quantities $pos$ and $neg$ of positive and negative roots of $Q$ satisfy Descartes' rule of signs. A couple $(σ_0,(pos,neg))$, where $σ_0$ is a sign pattern of length $d+1$, is {\em realizable} if there exists a polynomial $Q$ with $pos$ positive and $neg$ negative simple roots, with $(d-pos-neg)/2$ complex conjugate pairs and with $σ(Q)=σ_0$. We present a series of couples (sign pattern, pair $(pos,neg)$) depending on two integer parameters and with $pos\geq 1$, $neg\geq 1$, which is not realizable. For $d=9$, we give the exhaustive list of realizable couples with two sign changes in the sign pattern. △ Less

Submitted 9 February, 2023; originally announced February 2023.

Journal ref: Comptes Rendus, Mathématique Volume 362 (2024), p. 863-881

A domain free of the zeros of the partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We prove that for $q\in (0,1)$, the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ has no zeros in the closed domain $\{ \{ |x|\leq 3\} \cap \{${\rm Re}$x\leq 0\} \cap \{ |${\rm Im}$x|\leq 3/\sqrt{2}\} \} \subset \mathbb{C}$ and no real zeros $\geq -5$. We prove that for $q\in (0,1)$, the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ has no zeros in the closed domain $\{ \{ |x|\leq 3\} \cap \{${\rm Re}$x\leq 0\} \cap \{ |${\rm Im}$x|\leq 3/\sqrt{2}\} \} \subset \mathbb{C}$ and no real zeros $\geq -5$. △ Less

Submitted 28 October, 2022; originally announced October 2022.

Comments: 2 figures

MSC Class: 26A06

Journal ref: Mat. Stud. 58 (2022), no. 2, 142-158

No zeros of the partial theta function in the unit disk

Authors: Vladimir Petrov Kostov

Abstract: We prove that for $q\in (-1,0)\cup (0,1)$, the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ has no zeros in the closed unit disk. We prove that for $q\in (-1,0)\cup (0,1)$, the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ has no zeros in the closed unit disk. △ Less

Submitted 19 August, 2022; originally announced August 2022.

Journal ref: Annual of Sofia University "St. Kliment Ohridski'', Faculty of Mathematics and Informatics 111 (2024) 129-137

The disconnectedness of certain sets defined after uni-variate polynomials

Authors: Vladimir Petrov Kostov

Abstract: We consider the set of monic real univariate polynomials of a given degree $d$ with non-vanishing coefficients, with given signs of the coefficients and with given quantities $pos$ of their positive and $neg$ of their negative roots (all roots are distinct). For $d=6$ and for signs of the coefficients $(+,-,+,+,+,-,+)$, we prove that the set of such polynomials having two positive, two negative an… ▽ More We consider the set of monic real univariate polynomials of a given degree $d$ with non-vanishing coefficients, with given signs of the coefficients and with given quantities $pos$ of their positive and $neg$ of their negative roots (all roots are distinct). For $d=6$ and for signs of the coefficients $(+,-,+,+,+,-,+)$, we prove that the set of such polynomials having two positive, two negative and two complex conjugate roots, is not connected. For $pos+neg\leq 3$ and for any $d$, we give the exhaustive answer to the question for which signs of the coefficients there exist polynomials with such values of $pos$ and $neg$. △ Less

Submitted 22 August, 2022; v1 submitted 15 September, 2021; originally announced September 2021.

MSC Class: 26C10; 30C15

Journal ref: Constructive Mathematical Analysis, vol. 5 issue 3 (2022) 119-133

Descartes' rule of signs, canonical sign patterns and rigid orders of moduli

Authors: Vladimir Petrov Kostov

Abstract: We consider real polynomials in one variable without vanishing coefficients and with all roots real and of distinct moduli. We show that the signs of the coefficients define the order of the moduli of the roots on the real positive half-line exactly when no four consecutive signs of coefficients equal $(+,+,-,-)$, $(-,-,+,+)$, $(+,-,-,+)$ or $(-,+,+,-)$. We consider real polynomials in one variable without vanishing coefficients and with all roots real and of distinct moduli. We show that the signs of the coefficients define the order of the moduli of the roots on the real positive half-line exactly when no four consecutive signs of coefficients equal $(+,+,-,-)$, $(-,-,+,+)$, $(+,-,-,+)$ or $(-,+,+,-)$. △ Less

Submitted 8 March, 2021; originally announced March 2021.

Journal ref: Which Sign Patterns are Canonical?. Results Math 77 (2022) No 6, paper 235

Univariate polynomials and the contractibility of certain sets

Authors: Vladimir Petrov Kostov

Abstract: We consider the set $Π^*_d$ of monic polynomials $Q_d=x^d+\sum _{j=0}^{d-1}a_jx^j$, $x\in \mathbb{R}$, $a_j\in \mathbb{R}^*$, having $d$ distinct real roots, and its subsets defined by fixing the signs of the coefficients $a_j$. We show that for every choice of these signs, the corresponding subset is non-empty and contractible. A similar result holds true in the cases of polynomials $Q_d$ of even… ▽ More We consider the set $Π^*_d$ of monic polynomials $Q_d=x^d+\sum _{j=0}^{d-1}a_jx^j$, $x\in \mathbb{R}$, $a_j\in \mathbb{R}^*$, having $d$ distinct real roots, and its subsets defined by fixing the signs of the coefficients $a_j$. We show that for every choice of these signs, the corresponding subset is non-empty and contractible. A similar result holds true in the cases of polynomials $Q_d$ of even degree $d$ and having no real roots or of odd degree and having exactly one real root. For even $d$ and when $Q_d$ has exactly two real roots which are of opposite signs, the subset is contractible. For even $d$ and when $Q_d$ has two positive (resp. two negative) roots, the subset is contractible or empty. It is empty exactly when the constant term is positive, among the other even coefficients there is at least one which is negative, and all odd coefficients are positive (resp. negative). △ Less

Submitted 17 December, 2021; v1 submitted 21 January, 2021; originally announced January 2021.

Comments: 21 pages, 1 figure, to appear

Journal ref: Annual of Sofia University 107 (2020), 11-35

Sign patterns and rigid moduli orders

Authors: Yousra Gati, Vladimir Petrov Kostov, Mohamed Chaouki Tarchi

Abstract: We consider the set of monic degree $d$ real univariate polynomials $Q_d=x^d+\sum_{j=0}^{d-1}a_jx^j$ and its {\em hyperbolicity domain} $Π_d$, i.e. the subset of values of the coefficients $a_j$ for which the polynomial $Q_d$ has all roots real. The subset $E_d\subset Π_d$ is the one on which a modulus of a negative root of $Q_d$ is equal to a positive root of $Q_d$. At a point, where $Q_d$ has… ▽ More We consider the set of monic degree $d$ real univariate polynomials $Q_d=x^d+\sum_{j=0}^{d-1}a_jx^j$ and its {\em hyperbolicity domain} $Π_d$, i.e. the subset of values of the coefficients $a_j$ for which the polynomial $Q_d$ has all roots real. The subset $E_d\subset Π_d$ is the one on which a modulus of a negative root of $Q_d$ is equal to a positive root of $Q_d$. At a point, where $Q_d$ has $d$ distinct roots with exactly $s$ ($1\leq s\leq [d/2]$) equalities between positive roots and moduli of negative roots, the set $E_d$ is locally the transversal intersection of $s$ smooth hypersurfaces. At a point, where $Q_d$ has two double opposite roots and no other equalities between moduli of roots, the set $E_d$ is locally the direct product of $\mathbb{R}^{d-3}$ and a hypersurface in $\mathbb{R}^3$ having a Whitney umbrella singularity. For $d\leq 4$, we draw pictures of the sets $Π_d$ and~$E_d$. △ Less

Submitted 8 December, 2020; originally announced December 2020.

Journal ref: The Graduate Journal of Mathematics, Volume 6, Issue 1 (2021), 60-72

Hyperbolic polynomials and rigid moduli orders

Authors: Vladimir Petrov Kostov

Abstract: A hyperbolic polynomial (HP) is a real univariate polynomial with all roots real. By Descartes' rule of signs a HP with all coefficients nonvanishing has exactly $c$ positive and exactly $p$ negative roots counted with multiplicity, where $c$ and $p$ are the numbers of sign changes and sign preservations in the sequence of its coefficients. We consider HPs with distinct moduli of the roots. We ask… ▽ More A hyperbolic polynomial (HP) is a real univariate polynomial with all roots real. By Descartes' rule of signs a HP with all coefficients nonvanishing has exactly $c$ positive and exactly $p$ negative roots counted with multiplicity, where $c$ and $p$ are the numbers of sign changes and sign preservations in the sequence of its coefficients. We consider HPs with distinct moduli of the roots. We ask the question when the order of the moduli of the negative roots w.r.t. the positive roots on the real positive half-line completely determines the signs of the coefficients of the polynomial. When there is at least one positive and at least one negative root this is possible exactly when the moduli of the negative roots interlace with the positive roots (hence half or about half of the roots are positive). In this case the signs of the coefficients of the HP are either $(+,+,-,-,+,+,-,-,\ldots )$ or $(+,-,-,+,+,-,-,+,\ldots )$. △ Less

Submitted 26 August, 2020; originally announced August 2020.

Journal ref: Publicationes Mathematicae Debrecen 100/1-2 (2022), 119-128

Hyperbolic polynomials and canonical sign patterns

Authors: Vladimir Petrov Kostov

Abstract: A real univariate polynomial is hyperbolic if all its roots are real. By Descartes' rule of signs a hyperbolic polynomial (HP) with all coefficients nonvanishing has exactly $c$ positive and exactly $p$ negative roots counted with multiplicity, where $c$ and $p$ are the numbers of sign changes and sign preservations in the sequence of its coefficients. We discuss the question: If the moduli of all… ▽ More A real univariate polynomial is hyperbolic if all its roots are real. By Descartes' rule of signs a hyperbolic polynomial (HP) with all coefficients nonvanishing has exactly $c$ positive and exactly $p$ negative roots counted with multiplicity, where $c$ and $p$ are the numbers of sign changes and sign preservations in the sequence of its coefficients. We discuss the question: If the moduli of all $c+p$ roots are distinct and ordered on the positive half-axis, then at which positions can the $p$ moduli of negative roots be depending on the positions of the positive and negative signs of the coefficients of the polynomial? We are especially interested in the choices of these signs for which exactly one order of the moduli of the roots is possible. △ Less

Submitted 25 June, 2020; originally announced June 2020.

Journal ref: Serdica Math. J. 46 (2020) 135-150

Degree 5 polynomials and Descartes' rule of signs

Authors: Hassen Cheriha, Yousra Gati, Vladimir Petrov Kostov

Abstract: For a univariate real polynomial without zero coefficients, Descartes' rule of signs (completed by an observation of Fourier) says that its numbers $pos$ of positive and $neg$ of negative roots (counted with multiplicity) are majorized respectively by the numbers $c$ and $p$ of sign changes and sign preservartions in the sequence of its coefficients, and that the differences $c-pos$ and $p-neg$ ar… ▽ More For a univariate real polynomial without zero coefficients, Descartes' rule of signs (completed by an observation of Fourier) says that its numbers $pos$ of positive and $neg$ of negative roots (counted with multiplicity) are majorized respectively by the numbers $c$ and $p$ of sign changes and sign preservartions in the sequence of its coefficients, and that the differences $c-pos$ and $p-neg$ are even numbers. For degree 5 polynomials, it has been proved by A.~Albouy and Y.~Fu that there exist no such polynomials having three distinct positive and no negative roots and whose signs of the coefficients are $(+,+,-,+,-,-)$ (or having three distinct negative and no positive roots and whose signs of the coefficients are $(+,-,-,-,-,+)$). For degree 5 and when the leading coefficient is positive, these are all cases of numbers of positive and negative roots (all distinct) and signs of the coefficients which are compatible with Descartes' rule of signs, but for which there exist no such polynomials. We explain this non-existence and the existence in all other cases with $d=5$ by means of pictures showing the discriminant set of the family of polynomials $x^5+x^4+ax^3+bx^2+cx+d$ together with the coordinate axes. △ Less

Submitted 5 May, 2020; originally announced May 2020.

Journal ref: Acta Univ. M. Belii Ser. Math. 28 (2020), 3-21

A non-realization theorem in the context of Descartes' rule of signs

Authors: Hassen Cheriha, Yousra Gati, Vladimir Petrov Kostov

Abstract: For a real degree $d$ polynomial $P$ with all nonvanishing coefficients, with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients ($c+p=d$), Descartes' rule of signs says that $P$ has $pos\leq c$ positive and $neg\leq p$ negative roots, where $pos\equiv c($\, mod $2)$ and $neg\equiv p($\, mod $2)$. For $1\leq d\leq 3$, for every possible choice of the sequence of signs… ▽ More For a real degree $d$ polynomial $P$ with all nonvanishing coefficients, with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients ($c+p=d$), Descartes' rule of signs says that $P$ has $pos\leq c$ positive and $neg\leq p$ negative roots, where $pos\equiv c($\, mod $2)$ and $neg\equiv p($\, mod $2)$. For $1\leq d\leq 3$, for every possible choice of the sequence of signs of coefficients of $P$ (called sign pattern) and for every pair $(pos, neg)$ satisfying these conditions there exists a polynomial $P$ with exactly $pos$ positive and $neg$ negative roots (all of them simple); that is, all these cases are realizable. This is not true for $d\geq 4$, yet for $4\leq d\leq 8$ (for these degrees the exhaustive answer to the question of realizability is known) in all nonrealizable cases either $pos=0$ or $neg=0$. It was conjectured that this is the case for any $d\geq 4$. For $d=9$, we show a counterexample to this conjecture: for the sign pattern $(+,-,-,-,-,+,+,+,+,-)$ and the pair $(1,6)$ there exists no polynomial with $1$ positive, $6$ negative simple roots and a complex conjugate pairs and, up to equivalence, this is the only case for $d=9$. △ Less

Submitted 26 November, 2019; originally announced November 2019.

Comments: arXiv admin note: substantial text overlap with arXiv:1708.05530

Journal ref: Annual of Sofia University "St. Kliment Ohridski'', Faculty of Mathematics and Informatics vol. 106 (2019) 25-51

Two properties of the partial theta function

Authors: Vladimir Petrov Kostov

Abstract: For the partial theta function $θ(q,z):=\sum_{j=0}^{\infty}q^{j(j+1)/2}z^j$, $q$, $z\in \mathbb{C}$, $|q|<1$, we prove that its zero set is connected. This set is smooth at every point $(q^{\flat},z^{\flat})$ such that $z^{\flat}$ is a simple or double zero of $θ(q^{\flat},.)$. For $q\in (0,1)$, $q\rightarrow 1^-$ and $a\geq e^π$, there are $o(1/(1-q))$ and $(\ln (a/e^π))/(1-q)+o(1/(1-q))$ real ze… ▽ More For the partial theta function $θ(q,z):=\sum_{j=0}^{\infty}q^{j(j+1)/2}z^j$, $q$, $z\in \mathbb{C}$, $|q|<1$, we prove that its zero set is connected. This set is smooth at every point $(q^{\flat},z^{\flat})$ such that $z^{\flat}$ is a simple or double zero of $θ(q^{\flat},.)$. For $q\in (0,1)$, $q\rightarrow 1^-$ and $a\geq e^π$, there are $o(1/(1-q))$ and $(\ln (a/e^π))/(1-q)+o(1/(1-q))$ real zeros of $θ(q,.)$ in the intervals $[-e^π,0)$ and $[-a,-e^{-π}]$ respectively (and none in $[0,\infty)$). For $q\in (-1,0)$, $q\rightarrow -1^+$ and $a\geq e^{π/2}$, there are $o(1/(1+q))$ real zeros of $θ(q,.)$ in the interval $[-e^{π/2},e^{π/2}]$ and $(\ln (a/e^{π/2})/2)/(1+q)+o(1/(1+q))$ in each of the intervals $[-a,-e^{π/2}]$ and $[e^{π/2},a]$. △ Less

Submitted 20 November, 2019; originally announced November 2019.

Moduli of roots of hyperbolic polynomials and Descartes' rule of signs

Authors: Vladimir Petrov Kostov

Abstract: A real univariate polynomial with all roots real is called hyperbolic. By Descartes' rule of signs for hyperbolic polynomials (HPs) with all coefficients nonvanishing, a HP with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients has exactly $c$ positive and $p$ negative roots. For $c=2$ and for degree $6$ HPs, we discuss the question: When the moduli of the $6$ roots o… ▽ More A real univariate polynomial with all roots real is called hyperbolic. By Descartes' rule of signs for hyperbolic polynomials (HPs) with all coefficients nonvanishing, a HP with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients has exactly $c$ positive and $p$ negative roots. For $c=2$ and for degree $6$ HPs, we discuss the question: When the moduli of the $6$ roots of a HP are arranged in the increasing order on the real half-line, at which positions can be the moduli of its two positive roots depending on the positions of the two sign changes in the sequence of coefficients? △ Less

Submitted 12 September, 2019; originally announced September 2019.

Journal ref: Constructive Theory of Functions, Sozopol 2019 (B. Draganov, K. Ivanov, G. Nikolov and R. Uluchev, Eds.), pp. 131-146 Prof. Marin Drinov Academic Publishing House, Sofia, 2020

On Descartes' rule of signs

Authors: Hassen Cheriha, Yousra Gati, Vladimir Petrov Kostov

Abstract: A sequence of $d+1$ signs $+$ and $-$ beginning with a $+$ is called a {\em sign pattern (SP)}. We say that the real polynomial $P:=x^d+\sum _{j=0}^{d-1}a_jx^j$, $a_j\neq 0$, defines the SP $σ:=(+$,sgn$(a_{d-1})$, $\ldots$, sgn$(a_0))$. By Descartes' rule of signs, for the quantity $pos$ of positive (resp. $neg$ of negative) roots of $P$, one has $pos\leq c$ (resp. $neg\leq p=d-c$), where $c$ and… ▽ More A sequence of $d+1$ signs $+$ and $-$ beginning with a $+$ is called a {\em sign pattern (SP)}. We say that the real polynomial $P:=x^d+\sum _{j=0}^{d-1}a_jx^j$, $a_j\neq 0$, defines the SP $σ:=(+$,sgn$(a_{d-1})$, $\ldots$, sgn$(a_0))$. By Descartes' rule of signs, for the quantity $pos$ of positive (resp. $neg$ of negative) roots of $P$, one has $pos\leq c$ (resp. $neg\leq p=d-c$), where $c$ and $p$ are the numbers of sign changes and sign preservations in $σ$; the numbers $c-pos$ and $p-neg$ are even. We say that $P$ realizes the SP $σ$ with the pair $(pos, neg)$. For SPs with $c=2$, we give some sufficient conditions for the (non)realizability of pairs $(pos, neg)$ of the form $(0,d-2k)$, $k=1$, $\ldots$, $[(d-2)/2]$. △ Less

Submitted 6 May, 2019; originally announced May 2019.

Journal ref: On Descartes' rule for polynomials with two variations of sign, Lith. Math. J. 60 (2020), no. 4, 456-469

Descartes' rule of signs and moduli of roots

Authors: Vladimir Petrov Kostov

Abstract: A hyperbolic polynomial (HP) is a real univariate polynomial with all roots real. By Descartes' rule of signs a HP with all coefficients nonvanishing has exactly $c$ positive and exactly $p$ negative roots counted with multiplicity, where $c$ and $p$ are the numbers of sign changes and sign preservations in the sequence of its coefficients. For $c=1$ and $2$, we discuss the question: When the modu… ▽ More A hyperbolic polynomial (HP) is a real univariate polynomial with all roots real. By Descartes' rule of signs a HP with all coefficients nonvanishing has exactly $c$ positive and exactly $p$ negative roots counted with multiplicity, where $c$ and $p$ are the numbers of sign changes and sign preservations in the sequence of its coefficients. For $c=1$ and $2$, we discuss the question: When the moduli of all the roots of a HP are arranged in the increasing order on the real half-line, at which positions can be the moduli of its positive roots depending on the positions of the sign changes in the sequence of coefficients? △ Less

Submitted 24 April, 2019; originally announced April 2019.

Journal ref: Publicationes Mathematicae Debrecen 96/1-2 (2020) 161-184

On the complex conjugate zeros of the partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We prove that 1) for any $q\in (0,1)$, all complex conjugate pairs of zeros of the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ belong to the set $\{$~Re\,$x\in (-5792.7,0),$~$|$Im\,$x|<132~\}$ $\cup$ $\{ ~|x|<18~\}$ and 2) for any $q\in (-1,0)$, they belong to the rectangle $\{$~$|$Re\,$x|< 364.2,$~$|$Im\,$x|<132~\}$. We prove that 1) for any $q\in (0,1)$, all complex conjugate pairs of zeros of the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ belong to the set $\{$~Re\,$x\in (-5792.7,0),$~$|$Im\,$x|<132~\}$ $\cup$ $\{ ~|x|<18~\}$ and 2) for any $q\in (-1,0)$, they belong to the rectangle $\{$~$|$Re\,$x|< 364.2,$~$|$Im\,$x|<132~\}$. △ Less

Submitted 5 February, 2019; originally announced February 2019.

Journal ref: Functional Analysis and Applications 2019, 53:2, 149-152 (translation from Funkcional'nyy Analiz i Ego Prilozheniya 2019, 53:2, 87-91)

New aspects of Descartes' rule of signs

Authors: Vladimir Petrov Kostov, Boris Zalmanovich Shapiro

Abstract: Below we summarize some new developments in the area of distribution of roots and signs of real univariate polynomials pioneered by R.~Descartes in the middle of the 17-th century. Below we summarize some new developments in the area of distribution of roots and signs of real univariate polynomials pioneered by R.~Descartes in the middle of the 17-th century. △ Less

Submitted 21 January, 2019; originally announced January 2019.

Comments: 16 pages

MSC Class: Primary 26C10; Secondary 30C15

Addendum: A separation in modulus property of the zeros of a partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We consider the partial theta function $θ(q,z):=\sum _{j=0}^{\infty}q^{j(j+1)/2}z^j$, where $z\in \mathbb{C}$ is a variable and $q\in \mathbb{C}$, $0<|q|<1$, is a parameter. Set $D(a):=\{ q\in \mathbb{C}$, $0<|q|\leq a$, $\arg (q)\in [π/2,3π/2]\}$. We show that for $k\in \mathbb{N}$ and $q\in D(0.55)$, there exists exactly one zero of $θ(q,.)$ (which is a simple one) in the open annulus… ▽ More We consider the partial theta function $θ(q,z):=\sum _{j=0}^{\infty}q^{j(j+1)/2}z^j$, where $z\in \mathbb{C}$ is a variable and $q\in \mathbb{C}$, $0<|q|<1$, is a parameter. Set $D(a):=\{ q\in \mathbb{C}$, $0<|q|\leq a$, $\arg (q)\in [π/2,3π/2]\}$. We show that for $k\in \mathbb{N}$ and $q\in D(0.55)$, there exists exactly one zero of $θ(q,.)$ (which is a simple one) in the open annulus $|q|^{-k+1/2}<z<|q|^{-k-1/2}$ (if $k\geq 2$) or in the punctured disk $0<z<|q|^{-3/2}$ (if $k=1$). For $k=1$, $4$, $5$, $6$, $\ldots$, this holds true for $q\in D(0.6)$ as well. △ Less

Submitted 6 December, 2018; originally announced December 2018.

Journal ref: Vietnam Journal of Mathematics 48:1 March 2020, 145-157

On the zero set of the partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We consider the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$, where $q\in (-1,0)\cup (0,1)$ and either $x\in \mathbb{R}$ or $x\in \mathbb{C}$. We prove that for $x\in \mathbb{R}$, in each of the two cases $q\in (-1,0)$ and $q\in (0,1)$, its zero set consists of countably-many smooth curves in the $(q,x)$-plane each of which (with the exception of one curve for… ▽ More We consider the partial theta function $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$, where $q\in (-1,0)\cup (0,1)$ and either $x\in \mathbb{R}$ or $x\in \mathbb{C}$. We prove that for $x\in \mathbb{R}$, in each of the two cases $q\in (-1,0)$ and $q\in (0,1)$, its zero set consists of countably-many smooth curves in the $(q,x)$-plane each of which (with the exception of one curve for $q\in (-1,0)$) has a single point with a tangent line parallel to the $x$-axis. These points define double zeros of the function $θ(q,.)$; their $x$-coordinates belong to the interval $[-38.83\ldots ,-e^{1.4}=4.05\ldots )$ for $q\in (0,1)$ and to the interval $(-13.29,23.65)$ for $q\in (-1,0)$. For $q\in (0,1)$, infinitely-many of the complex conjugate pairs of zeros to which the double zeros give rise cross the imaginary axis and then remain in the half-disk $\{ |x|<18$, Re\,$x>0\}$. For $q\in (-1,0)$, complex conjugate pairs do not cross the imaginary axis. △ Less

Submitted 4 July, 2018; originally announced July 2018.

Comments: 4 figures

Journal ref: Serdica Mathematical journal 45 (2019) 225-258

Descartes' rule of signs, Rolle's theorem and sequences of admissible pairs

Authors: Hassen Cheriha, Yousra Gati, Vladimir Petrov Kostov

Abstract: Given a real univariate degree $d$ polynomial $P$, the numbers $pos_k$ and $neg_k$ of positive and negative roots of $P^{(k)}$, $k=0$, $\ldots$, $d-1$, must be admissible, i.e. they must satisfy certain inequalities resulting from Rolle's theorem and from Descartes' rule of signs. For $1\leq d\leq 5$, we give the answer to the question for which admissible $d$-tuples of pairs $(pos_k$, $neg_k)$ th… ▽ More Given a real univariate degree $d$ polynomial $P$, the numbers $pos_k$ and $neg_k$ of positive and negative roots of $P^{(k)}$, $k=0$, $\ldots$, $d-1$, must be admissible, i.e. they must satisfy certain inequalities resulting from Rolle's theorem and from Descartes' rule of signs. For $1\leq d\leq 5$, we give the answer to the question for which admissible $d$-tuples of pairs $(pos_k$, $neg_k)$ there exist polynomials $P$ with all nonvanishing coefficients such that for $k=0$, $\ldots$, $d-1$, $P^{(k)}$ has exactly $pos_k$ positive and $neg_k$ negative roots all of which are simple. △ Less

Submitted 11 May, 2018; originally announced May 2018.

Journal ref: Studia Scientiarum Mathematicarum Hungarica 57:2 (2020) 165-186 June 2020

A domain containing all zeros of the partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We consider the partial theta function, i.e. the sum of the bivariate series $θ(q,z):=\sum_{j=0}^{\infty}q^{j(j+1)/2}z^j$ for $q\in (0,1)$, $z\in \mathbb{C}$. We show that for any value of the parameter $q\in (0,1)$ all zeros of the function $θ(q,.)$ belong to the domain $\{ {\rm Re}~z<0, |{\rm Im}~z|\leq 132\}$$\cup$$\{ {\rm Re}~z\geq 0, |z|\leq 18\}$. We consider the partial theta function, i.e. the sum of the bivariate series $θ(q,z):=\sum_{j=0}^{\infty}q^{j(j+1)/2}z^j$ for $q\in (0,1)$, $z\in \mathbb{C}$. We show that for any value of the parameter $q\in (0,1)$ all zeros of the function $θ(q,.)$ belong to the domain $\{ {\rm Re}~z<0, |{\rm Im}~z|\leq 132\}$$\cup$$\{ {\rm Re}~z\geq 0, |z|\leq 18\}$. △ Less

Submitted 4 October, 2017; originally announced October 2017.

MSC Class: 26A06

Journal ref: Publicationes Mathematicae Debrecen 93, No. 1-2 (2018) 189-203

Polynomials, sign patterns and Descartes' rule of signs

Authors: Vladimir Petrov Kostov

Abstract: By Descartes' rule of signs, a real degree $d$ polynomial $P$ with all nonvanishing coefficients, with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients ($c+p=d$) has $pos\leq c$ positive and $neg\leq p$ negative roots, where $pos\equiv c($\, mod $2)$ and $neg\equiv p($\, mod $2)$. For $1\leq d\leq 3$, for every possible choice of the sequence of signs of coefficients… ▽ More By Descartes' rule of signs, a real degree $d$ polynomial $P$ with all nonvanishing coefficients, with $c$ sign changes and $p$ sign preservations in the sequence of its coefficients ($c+p=d$) has $pos\leq c$ positive and $neg\leq p$ negative roots, where $pos\equiv c($\, mod $2)$ and $neg\equiv p($\, mod $2)$. For $1\leq d\leq 3$, for every possible choice of the sequence of signs of coefficients of $P$ (called sign pattern) and for every pair $(pos, neg)$ satisfying these conditions there exists a polynomial $P$ with exactly $pos$ positive and exactly $neg$ negative roots (all of them simple). For $d\geq 4$ this is not so. It was observed that for $4\leq d\leq 10$, in all nonrealizable cases either $pos=0$ or $neg=0$. It was conjectured that this is the case for any $d\geq 4$. We show a counterexample to this conjecture for $d=11$. Namely, we prove that for the sign pattern $(+,-,-,-,-,-,+,+,+,+,+,-)$ and the pair $(1,8)$ there exists no polynomial with $1$ positive, $8$ negative simple roots and a complex conjugate pair. △ Less

Submitted 18 August, 2017; originally announced August 2017.

Journal ref: Mathematica Bohemica 144 (2019), No. 1, 39-67

A separation in modulus property of the zeros of a partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We consider the partial theta function $θ(q,z):=\sum _{j=0}^{\infty}q^{j(j+1)/2}z^j$, where $z\in \mathbb{C}$ is a variable and $q\in \mathbb{C}$, $0<|q|<1$, is a parameter. Set $α_0~:=~\sqrt{3}/2π~=~0.2756644477\ldots$. We show that, for $n\geq 5$, for $|q|\leq 1-1/(α_0n)$ and for $k\geq n$ there exists a unique zero $ξ_k$ of $θ(q,.)$ satisfying the inequalities $|q|^{-k+1/2}<|ξ_k|<|q|^{-k-1/2}$;… ▽ More We consider the partial theta function $θ(q,z):=\sum _{j=0}^{\infty}q^{j(j+1)/2}z^j$, where $z\in \mathbb{C}$ is a variable and $q\in \mathbb{C}$, $0<|q|<1$, is a parameter. Set $α_0~:=~\sqrt{3}/2π~=~0.2756644477\ldots$. We show that, for $n\geq 5$, for $|q|\leq 1-1/(α_0n)$ and for $k\geq n$ there exists a unique zero $ξ_k$ of $θ(q,.)$ satisfying the inequalities $|q|^{-k+1/2}<|ξ_k|<|q|^{-k-1/2}$; all these zeros are simple ones. The moduli of the remaining $n-1$ zeros are $\leq |q|^{-n+1/2}$. A {\em spectral value} of $q$ is a value for which $θ(q,.)$ has a multiple zero. We prove the existence of the spectral values $0.4353184958\ldots \pm i\, 0.1230440086\ldots$ for which $θ$ has double zeros $-5.963\ldots \pm i\, 6.104\ldots$. △ Less

Submitted 6 April, 2017; originally announced April 2017.

Journal ref: Analysis Mathematica 44 (2018), no. 4, 501-519

Something You Always Wanted to Know About Real Polynomials (But Were Afraid to Ask)

Authors: Vladimir Petrov Kostov, Boris Shapiro

Abstract: The famous Descartes' rule of signs from 1637 giving an upper bound on the number of positive roots of a real univariate polynomials in terms of the number of sign changes of its coefficients, has been an indispensable source of inspiration for generations of mathematicians. Trying to extend and sharpen this rule, we consider below the set of all real univariate polynomials of a given degree, a gi… ▽ More The famous Descartes' rule of signs from 1637 giving an upper bound on the number of positive roots of a real univariate polynomials in terms of the number of sign changes of its coefficients, has been an indispensable source of inspiration for generations of mathematicians. Trying to extend and sharpen this rule, we consider below the set of all real univariate polynomials of a given degree, a given collection of signs of their coefficients, and a given number of positive and negative roots. In spite of the elementary definition of the main object of our study, it is a non-trivial question for which sign patterns and numbers of positive and negative roots the corresponding set is non-empty. The main result of the present paper is a discovery of a new infinite family of non-realizable combinations of sign patterns and the numbers of positive and negative roots. △ Less

Submitted 13 March, 2017; originally announced March 2017.

Comments: 9 pages, many pictures

MSC Class: Primary 26C10; Secondary 30C15

Journal ref: Acta Universitatis Matthiae Belii series Mathematics, Issue 2019, 1--11

On realizability of sign patterns by real polynomials

Authors: Vladimir Petrov Kostov

Abstract: The classical Descartes' rule of signs limits the number of positive roots of a real polynomial in one variable by the number of sign changes in the sequence of its coefficients. One can ask the question which pairs of nonnegative integers $(p,n)$, chosen in accordance with this rule and with some other natural conditions, can be the pairs of numbers of positive and negative roots of a real polyno… ▽ More The classical Descartes' rule of signs limits the number of positive roots of a real polynomial in one variable by the number of sign changes in the sequence of its coefficients. One can ask the question which pairs of nonnegative integers $(p,n)$, chosen in accordance with this rule and with some other natural conditions, can be the pairs of numbers of positive and negative roots of a real polynomial with prescribed signs of the coefficients. The paper solves this problem for degree $8$ polynomials. △ Less

Submitted 9 March, 2017; originally announced March 2017.

Journal ref: Czechoslovak Math. J. 68 (143) (2018), no. 3, 853-874

On higher-order discriminants

Authors: Vladimir Petrov Kostov

Abstract: For the family of polynomials in one variable $P:=x^n+a_1x^{n-1}+\cdots +a_n$, $n\geq 4$, we consider its higher-order discriminant sets $\{ \tilde{D}_m=0\}$, where $\tilde{D}_m:=$Res$(P,P^{(m)})$, $m=2$, $\ldots$, $n-2$, and their projections in the spaces of the variables $a^k:=(a_1,\ldots ,a_{k-1},a_{k+1},\ldots ,a_n)$. Set $P^{(m)}:=\sum _{j=0}^{n-m}c_ja_jx^{n-m-j}$,… ▽ More For the family of polynomials in one variable $P:=x^n+a_1x^{n-1}+\cdots +a_n$, $n\geq 4$, we consider its higher-order discriminant sets $\{ \tilde{D}_m=0\}$, where $\tilde{D}_m:=$Res$(P,P^{(m)})$, $m=2$, $\ldots$, $n-2$, and their projections in the spaces of the variables $a^k:=(a_1,\ldots ,a_{k-1},a_{k+1},\ldots ,a_n)$. Set $P^{(m)}:=\sum _{j=0}^{n-m}c_ja_jx^{n-m-j}$, $P_{m,k}:=c_kP-x^mP^{(m)}$. We show that Res$(\tilde{D}_m,\partial \tilde{D}_m/\partial a_k,a_k)= A_{m,k}B_{m,k}C_{m,k}^2$, where $A_{m,k}=a_n^{n-m-k}$, $B_{m,k}=$Res$(P_{m,k},P_{m,k}')$ if $1\leq k\leq n-m$ and $A_{m,k}=a_{n-m}^{n-k}$, $B_{m,k}=$Res$(P^{(m)},P^{(m+1)})$ if $n-m+1\leq k\leq n$. The equation $C_{m,k}=0$ defines the projection in the space of the variables $a^k$ of the closure of the set of values of $(a_1,\ldots ,a_n)$ for which $P$ and $P^{(m)}$ have two distinct roots in common. The polynomials $B_{m,k},C_{m,k}\in \mathbb{C}[a^k]$ are irreducible. The result is generalized to the case when $P^{(m)}$ is replaced by a polynomial $P_*:=\sum _{j=0}^{n-m}b_ja_jx^{n-m-j}$, $0\neq b_i\neq b_j\neq 0$ for $i\neq j$. △ Less

Submitted 27 February, 2017; originally announced February 2017.

Journal ref: Bull. Sci. Math. 160 (2020), 102842, 27 pp

A property of discriminants

Authors: Vladimir Petrov Kostov

Abstract: For the family $P:=x^n+a_1x^{n-1}+\cdots +a_n$ of complex polynomials in the variable $x$ we study its {\em discriminant} $R:=$Res$(P,P',x)$, $R\in \mathbb{C}[a]$, $a=(a_1,\ldots ,a_n)$. When $R$ is regarded as a polynomial in $a_k$, one can consider its discriminant $\tilde{D}_k:=$Res$(R,\partial R/\partial a_k,a_k)$. We show that $\tilde{D}_k=c_k(a_n)^{d(n,k)}M_k^2T_k^3$, where… ▽ More For the family $P:=x^n+a_1x^{n-1}+\cdots +a_n$ of complex polynomials in the variable $x$ we study its {\em discriminant} $R:=$Res$(P,P',x)$, $R\in \mathbb{C}[a]$, $a=(a_1,\ldots ,a_n)$. When $R$ is regarded as a polynomial in $a_k$, one can consider its discriminant $\tilde{D}_k:=$Res$(R,\partial R/\partial a_k,a_k)$. We show that $\tilde{D}_k=c_k(a_n)^{d(n,k)}M_k^2T_k^3$, where $c_k\in \mathbb{Q}^*$, $d(n,k):=\min (1,n-k)+\max (0,n-k-2)$, the polynomials $M_k,T_k\in \mathbb{C}[a^k]$ have integer coefficients, $a^k=(a_1,\ldots ,a_{k-1},a_{k+1},\ldots ,a_n)$, the sets $\{ M_k=0\}$ and $\{ T_k=0\}$ are the projections in the space of the variables $a^k$ of the closures of the strata of the variety $\{ R=0\}$ on which $P$ has respectively two double roots or a triple root. Set $P_k:=P-xP'/(n-k)$ for $1\leq k\leq n-1$ and $P_n:=P'$. One has $T_k=${\rm Res}$(P_k,P_k',x)$ for $k\neq n-1$ and $T_{n-1}=${\rm Res}$(P_{n-1},P_{n-1}',x)/a_n$. △ Less

Submitted 11 January, 2017; originally announced January 2017.

Journal ref: Vietnam Journal of Mathematics 47:2 (2019) 287-296

Some facts about discriminants

Authors: Vladimir Petrov Kostov

Abstract: For the family of polynomials in one variable $P:=x^n+a_1x^{n-1}+\cdots +a_n$ we ask the questions at which points its discriminant set can be considered as the graph of a function of all coefficients $a_j$ but one and how its subset of points, where the discriminant set is not smooth, projects on the different coordinate hyperplanes in the space of the coefficients~$a$. For the family of polynomials in one variable $P:=x^n+a_1x^{n-1}+\cdots +a_n$ we ask the questions at which points its discriminant set can be considered as the graph of a function of all coefficients $a_j$ but one and how its subset of points, where the discriminant set is not smooth, projects on the different coordinate hyperplanes in the space of the coefficients~$a$. △ Less

Submitted 21 December, 2016; originally announced December 2016.

Journal ref: Comptes Rendus Acad. Bulg. Sci. 70, No. 4 (2017) 467-476

Uniform bounds on locations of zeros of partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We consider the partial theta function $θ(q,z):=\sum _{j=0}^{\infty}q^{j(j+1)/2}z^j$, where $(q,z)\in \mathbb{C}^2$, $|q|<1$. We show that for any $0<δ_0<δ<1$, there exists $n_0\in \mathbb{N}$ such that for any $q$ with $δ_0\leq |q|\leq δ$ and for any $n\geq n_0$ the function $θ$ has exactly $n$ zeros with modulus $<|q|^{-n-1/2}$ counted with multiplicity. We consider the partial theta function $θ(q,z):=\sum _{j=0}^{\infty}q^{j(j+1)/2}z^j$, where $(q,z)\in \mathbb{C}^2$, $|q|<1$. We show that for any $0<δ_0<δ<1$, there exists $n_0\in \mathbb{N}$ such that for any $q$ with $δ_0\leq |q|\leq δ$ and for any $n\geq n_0$ the function $θ$ has exactly $n$ zeros with modulus $<|q|^{-n-1/2}$ counted with multiplicity. △ Less

Submitted 19 July, 2016; originally announced July 2016.

Journal ref: Acta Universitatis Matthiae Belii Ser. Math. (2016) 16-20

On the multiple zeros of a partial theta function

Authors: Vladimir Petrov Kostov

Abstract: We consider the partial theta function $θ(q,x):=\sum_{j=0}^{\infty}q^{j(j+1)/2}x^j$, where $x\in \mathbb{C}$ is a variable and $q\in \mathbb{C}$, $0<|q|<1$, is a parameter. We show that, for any fixed $q$, if $ζ$ is a multiple zero of the function $θ(q,.)$, then $|ζ|\leq 8^{11}$. We consider the partial theta function $θ(q,x):=\sum_{j=0}^{\infty}q^{j(j+1)/2}x^j$, where $x\in \mathbb{C}$ is a variable and $q\in \mathbb{C}$, $0<|q|<1$, is a parameter. We show that, for any fixed $q$, if $ζ$ is a multiple zero of the function $θ(q,.)$, then $|ζ|\leq 8^{11}$. △ Less

Submitted 29 February, 2016; originally announced February 2016.

Journal ref: Funct. Anal. Appl. 50, No. 2 (2016) 153-156

On a partial theta function and its spectrum

Authors: Vladimir Petrov Kostov

Abstract: The bivariate series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ %(where $(q,x)\in {\bf C}^2$, $|q|<1$) defines a {\em partial theta function}. For fixed $q$ ($|q|<1$), $θ(q,.)$ is an entire function. For $q\in (-1,0)$ the function $θ(q,.)$ has infinitely many negative and infinitely many positive real zeros. There exists a sequence $\{ \bar{q}_j\}$ of values of $q$ tending to $-1^+$ such that… ▽ More The bivariate series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ %(where $(q,x)\in {\bf C}^2$, $|q|<1$) defines a {\em partial theta function}. For fixed $q$ ($|q|<1$), $θ(q,.)$ is an entire function. For $q\in (-1,0)$ the function $θ(q,.)$ has infinitely many negative and infinitely many positive real zeros. There exists a sequence $\{ \bar{q}_j\}$ of values of $q$ tending to $-1^+$ such that $θ(\bar{q}_k,.)$ has a double real zero $\bar{y}_k$ (the rest of its real zeros being simple). For $k$ odd (resp. for $k$ even) $θ(\bar{q}_k,.)$ has a local minimum at $\bar{y}_k$ and $\bar{y}_k$ is the rightmost of the real negative zeros of $θ(\bar{q}_k,.)$ (resp. $θ(\bar{q}_k,.)$ has a local maximum at $\bar{y}_k$ and for $k$ sufficiently large $\bar{y}_k$ is the second from the left of the real negative zeros of $θ(\bar{q}_k,.)$). For $k$ sufficiently large one has $-1<\bar{q}_{k+1}<\bar{q}_k<0$. One has $\bar{q}_k=1-(π/8k)+o(1/k)$ and $|\bar{y}_k|\rightarrow e^{π/2}=4.810477382\ldots$. △ Less

Submitted 22 April, 2015; originally announced April 2015.

Comments: 4 figures

Journal ref: Proc. Royal Soc. Edinb. A 146, No. 3 (2016) 609-623

On the double zeros of a partial theta function

Authors: Vladimir Petrov Kostov

Abstract: The series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ converges for $q\in [0,1)$, $x\in \mathbb{R}$, and defines a {\em partial theta function}. For any fixed $q\in (0,1)$ it has infinitely many negative zeros. For $q$ taking one of the {\em spectral} values $\tilde{q}_1$, $\tilde{q}_2$, $\ldots$ (where $0.3092493386\ldots =\tilde{q}_1<\tilde{q}_2<\cdots <1$,… ▽ More The series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ converges for $q\in [0,1)$, $x\in \mathbb{R}$, and defines a {\em partial theta function}. For any fixed $q\in (0,1)$ it has infinitely many negative zeros. For $q$ taking one of the {\em spectral} values $\tilde{q}_1$, $\tilde{q}_2$, $\ldots$ (where $0.3092493386\ldots =\tilde{q}_1<\tilde{q}_2<\cdots <1$, $\lim _{j\rightarrow \infty}\tilde{q}_j=1$) the function $θ(q,.)$ has a double zero $y_j$ which is the rightmost of its real zeros (the rest of them being simple). For $q\neq \tilde{q}_j$ the partial theta function has no multiple real zeros. We prove that $\tilde{q}_j=1-π/2j+(\log j)/8j^2+O(1/j^2)$ and $y_j=-e^πe^{-(\log j)/4j+O(1/j)}$. △ Less

Submitted 22 April, 2015; originally announced April 2015.

Comments: 1 figure

Journal ref: Bull. Sci. Math. 140, No. 4 (2016) 98-111

Interlacing properties and the Schur-Szegő composition

Authors: Vladimir Petrov Kostov

Abstract: Each degree $n$ polynomial in one variable of the form $(x+1)(x^{n-1}+c_1x^{n-2}+\cdots +c_{n-1})$ is representable in a unique way as a Schur-Szegő composition of $n-1$ polynomials of the form $(x+1)^{n-1}(x+a_i)$, see \cite{Ko1}, \cite{AlKo} and \cite{Ko2}. Set $σ_j:=\sum _{1\leq i_1<\cdots <i_j\leq n-1}a_{i_1}\cdots a_{i_j}$. The eigenvalues of the affine mapping… ▽ More Each degree $n$ polynomial in one variable of the form $(x+1)(x^{n-1}+c_1x^{n-2}+\cdots +c_{n-1})$ is representable in a unique way as a Schur-Szegő composition of $n-1$ polynomials of the form $(x+1)^{n-1}(x+a_i)$, see \cite{Ko1}, \cite{AlKo} and \cite{Ko2}. Set $σ_j:=\sum _{1\leq i_1<\cdots <i_j\leq n-1}a_{i_1}\cdots a_{i_j}$. The eigenvalues of the affine mapping $(c_1,\ldots ,c_{n-1})\mapsto (σ_1,\ldots ,σ_{n-1})$ are positive rational numbers and its eigenvectors are defined by hyperbolic polynomials (i.e. with real roots only). In the present paper we prove interlacing properties of the roots of these polynomials. △ Less

Submitted 9 April, 2015; originally announced April 2015.

Journal ref: Functional Analysis and Other Mathematics Volume 3, Issue 1 (2010), 65-74

A mapping defined by the Schur-Szegő composition

Authors: Vladimir Petrov Kostov

Abstract: Each degree $n+k$ polynomial of the form $(x+1)^k(x^n+c_1x^{n-1}+\cdots +c_n)$, $k\in \mathbb{N}$, is representable as Schur-Szegő composition of $n$ polynomials of the form $(x+1)^{n+k-1}(x+a_j)$. We study properties of the affine mapping $Φ_{n,k}$~:~$(c_1,\ldots ,c_n)$ $\mapsto$ $(σ_1, \ldots ,σ_n)$, where $σ_i$ are the elementary symmetric polynomials of the numbers $a_j$. We study also propert… ▽ More Each degree $n+k$ polynomial of the form $(x+1)^k(x^n+c_1x^{n-1}+\cdots +c_n)$, $k\in \mathbb{N}$, is representable as Schur-Szegő composition of $n$ polynomials of the form $(x+1)^{n+k-1}(x+a_j)$. We study properties of the affine mapping $Φ_{n,k}$~:~$(c_1,\ldots ,c_n)$ $\mapsto$ $(σ_1, \ldots ,σ_n)$, where $σ_i$ are the elementary symmetric polynomials of the numbers $a_j$. We study also properties of a similar mapping for functions of the form $e^xP$, where $P$ is a polynomial, $P(0)=1$, and we extend the Descartes rule to them. △ Less

Submitted 8 April, 2015; originally announced April 2015.

Journal ref: Comptes Rendus Acad. Sci. Bulgare Vol. 63 (2010) No. 7 943 -- 952

A refined realization theorem in the context of the Schur-Szegő composition

Authors: Vladimir Petrov Kostov

Abstract: Every polynomial of the form $P=(x+1)(x^{n-1}+c_1x^{n-2}+\cdots +c_{n-1})$ is representable as Schur-Szegő composition of $n-1$ polynomials of the form $(x+1)^{n-1}(x+a_i)$, where the numbers $a_i$ are unique up to permutation. We give necessary and sufficient conditions upon the possible values of the $8$-vector whose components are the number of positive, zero, negative and complex roots of a re… ▽ More Every polynomial of the form $P=(x+1)(x^{n-1}+c_1x^{n-2}+\cdots +c_{n-1})$ is representable as Schur-Szegő composition of $n-1$ polynomials of the form $(x+1)^{n-1}(x+a_i)$, where the numbers $a_i$ are unique up to permutation. We give necessary and sufficient conditions upon the possible values of the $8$-vector whose components are the number of positive, zero, negative and complex roots of a real polynomial $P$ and the number of positive, zero, negative and complex among the quantities $a_i$ corresponding to $P$. A similar result is proved about entire functions of the form $e^xR$, where $R$ is a polynomial. △ Less

Submitted 7 April, 2015; originally announced April 2015.

Journal ref: Bull. Sci. Math. 136, No. 5 (2012) 507-520

A property of a partial theta function

Authors: Vladimir Petrov Kostov

Abstract: The series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ converges for $|q|<1$ and defines a {\em partial theta function}. For any fixed $q\in (0,1)$ it has infinitely many negative zeros. It is known that for $q$ taking one of the {\em spectral} values $\tilde{q}_1$, $\tilde{q}_2$, $\ldots$ (where $0.3092493386\ldots =\tilde{q}_1<\tilde{q}_2<\cdots <1$, $\lim _{j\rightarrow \infty}\tilde{q}_j=1$)… ▽ More The series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ converges for $|q|<1$ and defines a {\em partial theta function}. For any fixed $q\in (0,1)$ it has infinitely many negative zeros. It is known that for $q$ taking one of the {\em spectral} values $\tilde{q}_1$, $\tilde{q}_2$, $\ldots$ (where $0.3092493386\ldots =\tilde{q}_1<\tilde{q}_2<\cdots <1$, $\lim _{j\rightarrow \infty}\tilde{q}_j=1$) the function $θ(q,.)$ has a double zero which is the rightmost of its real zeros (the rest of them being simple). For $q\neq \tilde{q}_j$ the partial theta function has no multiple real zeros. We prove that: 1) for $q\in (\tilde{q}_{j},\tilde{q}_{j+1}]$ the function $θ$ is a product of a degree $2j$ real polynomial without real roots and a function of the Laguerre-Pólya class $\cal{LP-I}$; 2) for $q\in \mathbb{C}\backslash 0$, $|q|<1$, $θ(q,x)=\prod _i(1+x/x_i)$, where $-x_i$ are the zeros of $θ$; 3) for any fixed $q\in \mathbb{C}\backslash 0$, $|q|<1$, the function $θ$ has at most finitely-many multiple zeros; 4) for any $q\in (-1,0)$ the function $θ$ is a product of a real polynomial without real zeros and a function of the Laguerre-Pólya class $\cal{LP}$. 5) for any fixed $q\in \mathbb{C}\backslash 0$, $|q|<1$, and for $k$ sufficiently large, the function $θ$ has a zero $ζ_k$ close to $-q^{-k}$. These are all but finitely-many of the zeros of $θ$. △ Less

Submitted 7 April, 2015; originally announced April 2015.

Journal ref: Comptes Rendus Acad. Sci. Bulgare 67, No. 10 (2014) 1319-1326

On the spectrum of a partial theta function

Authors: Vladimir Petrov Kostov

Abstract: The bivariate series $θ(q,x):=\sum_{j=0}^{\infty}q^{j(j+1)/2}x^j$ defines a {\em partial theta function}. For fixed $q$, $θ(q,.)$ is an entire function. We show that for $|q|\leq 0.108$ the function $θ(q,.)$ has no multiple zeros. The bivariate series $θ(q,x):=\sum_{j=0}^{\infty}q^{j(j+1)/2}x^j$ defines a {\em partial theta function}. For fixed $q$, $θ(q,.)$ is an entire function. We show that for $|q|\leq 0.108$ the function $θ(q,.)$ has no multiple zeros. △ Less

Submitted 3 April, 2015; originally announced April 2015.

Journal ref: Proc. Roy. Soc. Edinb. Sect. A 144 (2014), no. 5, 925-933

Asymptotic expansions of zeros of a partial theta function

Authors: Vladimir Petrov Kostov

Abstract: The bivariate series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ defines a {\em partial theta function}. For fixed $q$ ($|q|<1$), $θ(q,.)$ is an entire function. We prove a property of stabilization of the coefficients of the Laurent series in $q$ of the zeros of $θ$. The coefficients $r_k$ of the stabilized series are positive integers. They are the elements of a known increasing sequence satis… ▽ More The bivariate series $θ(q,x):=\sum _{j=0}^{\infty}q^{j(j+1)/2}x^j$ defines a {\em partial theta function}. For fixed $q$ ($|q|<1$), $θ(q,.)$ is an entire function. We prove a property of stabilization of the coefficients of the Laurent series in $q$ of the zeros of $θ$. The coefficients $r_k$ of the stabilized series are positive integers. They are the elements of a known increasing sequence satisfying the recurrence relation $r_k=\sum _{ν=1}^{\infty}(-1)^{ν-1}(2ν+1)r_{k-ν(ν+1)/2}$. △ Less

Submitted 3 April, 2015; originally announced April 2015.

Journal ref: C. R. Acad. Bulgare Sci. 68 (2015), no. 4, 419-426

Could René Descartes have known this?

Authors: Jens Forsgard, Vladimir P. Kostov, Boris Shapiro

Abstract: Below we discuss the partition of the space of real univariate polynomials according to the number of positive and negative roots and signs of the coefficients. We present several series of non-realizable combinations of signs together with the numbers of positive and negative roots. We provide a detailed information about possible non-realizable combinations as above up to degree 8 as well as a g… ▽ More Below we discuss the partition of the space of real univariate polynomials according to the number of positive and negative roots and signs of the coefficients. We present several series of non-realizable combinations of signs together with the numbers of positive and negative roots. We provide a detailed information about possible non-realizable combinations as above up to degree 8 as well as a general conjecture about such combinations. △ Less

Submitted 5 January, 2015; originally announced January 2015.

Comments: 15 pages, no figures

MSC Class: Primary 26C10; Secondary 30C15

Hardy-Petrovitch-Hutchinson's problem and partial theta function

Authors: Vladimir Petrov Kostov, Boris Shapiro

Abstract: In 1907 M.Petrovitch initiated the study of a class of entire functions all whose finite sections are real-rooted polynomials. An explicit description of this class in terms of the coefficients of a series is impossible since it is determined by an infinite number of discriminantal inequalities one for each degree. However, interesting necessary or sufficient conditions can be formulated. In parti… ▽ More In 1907 M.Petrovitch initiated the study of a class of entire functions all whose finite sections are real-rooted polynomials. An explicit description of this class in terms of the coefficients of a series is impossible since it is determined by an infinite number of discriminantal inequalities one for each degree. However, interesting necessary or sufficient conditions can be formulated. In particular, J.I.Hutchinson has shown that an entire function p(x)=a_0+a_1x+...+a_nx^n+... with strictly positive coefficients has the property that any its finite segment a_ix^i+...+a_jx^j has all real roots if and only if for all i=1,2,... one has a_i^2/a_{i-1}a_{i+1} is greater than or equal to 4. In the present paper we give sharp lower bounds on the ratios a_i^2/a_{i-1}a_{i+1} for the class considered by M.Petrovitch. In particular, we show that the limit of these minima when i tends to infinity equals the inverse of the maximal positive value of the parameter for which the classical partial theta function belongs to the Laguerre-Polya class. △ Less

Submitted 30 June, 2011; originally announced June 2011.

MSC Class: Primary 30C15; Secondary 12D10; 26C05

Journal ref: Duke Math. J. 162, no. 5 (2013), 825-861

On the Deligne-Simpson problem and its weak version

Authors: Vladimir Petrov Kostov

Abstract: We consider the {\em Deligne-Simpson problem (DSP) (resp. the weak DSP): Give necessary and sufficient conditions upon the choice of the $p+1$ conjugacy classes $c_j\subset gl(n,{\bf C})$ or $C_j\subset GL(n,{\bf C})$ so that there exist irreducible $(p+1)$-tuples (resp. $(p+1)$-tuples with trivial centralizers) of matrices $A_j\in c_j$ with zero sum or of matrices $M_j\in C_j$ whose product is… ▽ More We consider the {\em Deligne-Simpson problem (DSP) (resp. the weak DSP): Give necessary and sufficient conditions upon the choice of the $p+1$ conjugacy classes $c_j\subset gl(n,{\bf C})$ or $C_j\subset GL(n,{\bf C})$ so that there exist irreducible $(p+1)$-tuples (resp. $(p+1)$-tuples with trivial centralizers) of matrices $A_j\in c_j$ with zero sum or of matrices $M_j\in C_j$ whose product is $I$.} The matrices $A_j$ (resp. $M_j$) are interpreted as matrices-residua of Fuchsian linear systems (resp. as monodromy matrices of regular linear systems) of differential equations with complex time. In the paper we give sufficient conditions for solvability of the DSP in the case when one of the matrices is with distinct eigenvalues. △ Less

Submitted 28 October, 2003; originally announced October 2003.

Comments: 22 pages

MSC Class: 15A24

Journal ref: Bulletin des Sciences Mathematiques 128/2 (2004) 105-125

On arrangements of the roots of a hyperbolic polynomial and of one of its derivatives

Authors: Vladimir Petrov Kostov

Abstract: We consider real monic {\em hyperbolic} polynomials in one real variable, i.e. polynomials having only real roots. Call {\em hyperbolicity domain} $Π$ of the family of polynomials $P(x,a)=x^n+a_1x^{n-1}+... +a_n$, $a_i,x\in {\bf R}$, the set $\{a\in {\bf R}^n| P $is hyperbolic $\}$. The paper studies a stratification of $Π$ defined by the arrangement of the roots of $P$ and $P^{(k)}$, where… ▽ More We consider real monic {\em hyperbolic} polynomials in one real variable, i.e. polynomials having only real roots. Call {\em hyperbolicity domain} $Π$ of the family of polynomials $P(x,a)=x^n+a_1x^{n-1}+... +a_n$, $a_i,x\in {\bf R}$, the set $\{a\in {\bf R}^n| P $is hyperbolic $\}$. The paper studies a stratification of $Π$ defined by the arrangement of the roots of $P$ and $P^{(k)}$, where $2\leq k\leq n-1$. We prove that the strata are smooth contractible real algebraic varieties. △ Less

Submitted 7 November, 2002; originally announced November 2002.

On a stratification defined by real roots of polynomials

Authors: Vladimir Petrov Kostov

Abstract: We consider the family of polynomials $P(x,a)=x^n+a_1x^{n-1}+... +a_n$, $x,a_i\in {\bf R}$, and the stratification of ${\bf R}^n\cong \{(a_1,... ,a_n)|a_i\in {\bf R}\}$ defined by the multiplicity vector of the real roots of $P$. We prove smoothness of the strata and a transversality property of their tangent spaces. We consider the family of polynomials $P(x,a)=x^n+a_1x^{n-1}+... +a_n$, $x,a_i\in {\bf R}$, and the stratification of ${\bf R}^n\cong \{(a_1,... ,a_n)|a_i\in {\bf R}\}$ defined by the multiplicity vector of the real roots of $P$. We prove smoothness of the strata and a transversality property of their tangent spaces. △ Less

Submitted 2 September, 2002; v1 submitted 28 August, 2002; originally announced August 2002.

Journal ref: Serdica Math. J. 29, No. 2 (2003), 177-186