WorksMathematics
A reciprocal Schoenberg inequality and its extensions
Abstract
We study reciprocal moments of the critical points of a polynomial with a simple zero at the origin. A positive similarity transforms a reciprocal derivative companion into a complex symmetric matrix. This gives a short proof of the reciprocal Schoenberg inequality and a refinement for every elementary symmetric function. We determine the optimal constants in the corresponding inequalities for every positive exponent, including the sublinear range, and extend the results to arbitrary positive charges. We also characterize the normal powers of the companion and of its symmetric representative. The resulting criterion gives the complete equality cases in two quartic inequalities and explains the appearance of noncollinear equality configurations at higher orders. We also obtain a closed generating determinant and a quantitative collinearity estimate.
Introduction
The Gauss–Lucas theorem places the critical points of a polynomial in the convex hull of its zeros. Moment inequalities give a different kind of information: they compare the distribution of the critical points with that of the zeros. A distinguished example is Schoenberg’s inequality. If a polynomial of degree has zeros and critical points , then
Equality holds precisely when the zeros are collinear. Schoenberg proposed the centered form of this inequality in 1986 [9]. Pereira [8] and Malamud [7] proved it through matrix representations of differentiation. These representations also connect polynomial geometry with the majorization relations of de Bruijn and Springer.
Cheung and Ng [4] developed derivative companion matrices and used them to prove higher-order inequalities, including a conjecture of de Bruin and Sharma. Kushel and Tyaglov [6] obtained a sharper quartic inequality through circulant matrices. Borcea [3] extended the majorization framework to equilibrium points of arbitrary positive charge distributions. Exterior powers already play an essential role in this literature: they replace individual eigenvalues by products of eigenvalues.
Negative moments arise naturally when one measures distances from a distinguished zero. We therefore write
and denote the zeros of by . Since , all the are nonzero. Tang and Zhang [10 Conjecture 6.3] proposed the reciprocal inequality
Their Conjecture 1.13 asks for the optimal power-sum bound
We prove both inequalities, give their equality cases, and place them in a larger family. At , the equality assertion in Conjecture 1.13 requires a correction: the nonzero zeros need only lie on one ray, rather than coincide.
The basic observation is simple. Put and let be the all-ones matrix. The matrix has eigenvalues . Its similar representative
is complex symmetric and has a smaller Frobenius norm. Schur’s inequality applied to proves (3); applying it to exterior powers of retains the same collinearity equality condition. A second factorization removes the phases of the and leads, through the Araki–Lieb–Thirring inequality, to sharp constants for all positive exponents.
For unit charges, the phase comparison itself is contained in Cheung–Ng’s Theorem 2.1 after inversion and the use of the product identity. Likewise, the endpoint of our elementary symmetric bounds follows from Borcea’s product majorization. Our bounds for improve the coefficients supplied by that majorization, as explained in Remark 4.4. The additional phase-sensitive refinements and the equality analysis use the same companion, but require its symmetric representative. In particular, a divided-difference formula characterizes all its normal powers. This gives the complete equality classification left open in [10 Remark 7.2] and shows why the refined quartic inequality has only collinear equality cases, whereas higher even orders need not.
Definitions and notation
Throughout, are nonzero complex numbers and are positive real numbers. Repetitions among the are allowed. Define
Let be its zeros, counted with multiplicity, and put
The polynomial has degree , leading coefficient , and a nonzero constant term. Thus all these reciprocals are defined.
For distinct positions, the are the zeros of
the force function associated with positive charges . For repeated positions we use the polynomial definition (2.1), so that roots lost through rational cancellation are retained. Arbitrary positive charges are obtained by taking . The choice gives for (1.2). More generally, gives the nonzero critical points of , where is a positive integer.
Our matrices are
Here is the positive square root. The matrices , are real symmetric, , and . Lemma 4.1 identifies their spectrum with .
For a vector , let denote its th elementary symmetric function. We abbreviate
with . A sum over runs over all -element subsets of . The Frobenius and operator norms are denoted by and . We use for the conjugate transpose, for entrywise conjugation, and . Collinearity below always means collinearity on a line through the origin. It is equivalent for the configurations and .
Main results
Sharp inequalities for all positive exponents
Theorem 3.1. For and ,
For ,
If , equality in (3.1) holds exactly when all coincide for , and exactly when they have the same argument for . Inequality (3.2) is strict for , but each of its coefficients is individually optimal. For , both inequalities are identities for every .
Corollary 3.2. Suppose . The best constant in
is
For , the equality cases are those in Theorem 3.1; in the sublinear range the optimal constant is not attained. For , equality always holds.
Taking and proves (1.4). For example, attains equality at , although its nonzero zeros are distinct. At the other end, records the product identity
A refinement retaining the phases
Theorem 3.3. For , define
Then
For , equality holds if and only if the are collinear. For , equality always holds. Moreover,
For unit charges this reads
At it is precisely (3), including its complete equality characterization. Identity (13) measures the improvement over Theorem 3.1 at by a sum of squared pairwise distances.
Normal powers and even moments
Define monic polynomials of degree by
Thus , where .
Theorem 3.4. For every positive integer ,
and the second difference is the weighted variance
The following equality criteria hold:
(1) is normal if and only if .
(2) is normal if and only if the are collinear or .
Thus equality in the first inequality of (16) has exactly the configurations in (2), while equality between its first and last terms has exactly those in (1). Every noncollinear equality configuration has at most distinct nonzero zeros.
For the quartic case, introduce
Corollary 3.5. One has
Equality in the first inequality holds if and only if the are collinear. Equality between the first and last expressions holds if and only if either all coincide, or they take exactly two distinct values satisfying
For , the last expression in (19) is the bound in [10 Theorem 7.1]. Its equality cases, apart from coincident roots, are therefore
where and are nonzero. The opposite pair for was already noted in [10 Remark 7.2]; the displayed relation gives the complete classification.
The symmetric companion and the sharp inequalities
Representation and the quadratic inequality
Lemma 4.1. The matrices and in (6) have eigenvalues . The matrix is normal if and only if the are collinear, and
Proof. The rank-one determinant formula gives the polynomial identity
The expression on the right is monic and has precisely the required roots. This also proves (10). Since , it has the same spectrum.
Set . Direct multiplication gives
The entry of the middle matrix is . It vanishes for every pair exactly when the lie on one line through the origin. Finally, ; expanding gives (21).
Schur triangularization yields, for any square matrix ,
with equality exactly when is normal. Indeed, the difference is the sum of the squared moduli of the strictly upper triangular entries of a Schur form. Lemma 4.1 and (23) prove the weighted reciprocal Schoenberg inequality, and hence (3), with equality.
Exterior powers
Write . In the standard basis indexed by -element subsets, its entry is . The identities
follow from Cauchy–Binet. The eigenvalues of are the -fold products of the eigenvalues of ; its singular values are the corresponding products of singular values. For positive definite , functional calculus gives for real .
We will use the following elementary equality observation.
Lemma 4.2. If is invertible and , then is normal if and only if is normal.
Proof. It suffices to show that is injective on positive definite matrices. If , then . Every -fold product of the positive eigenvalues of is therefore . Comparing products that differ in one index shows that all the eigenvalues are equal; hence each is and . Apply this to and . □
Proof of Theorem 3.3. Apply (4.3) to . Its squared Frobenius norm is
The minors of have a particularly simple form:
Every other minor is zero, since two or more rows outside the common index set are proportional. Grouping the off-diagonal terms in (4.4) by gives exactly (3.5). Equality follows from Lemmas 4.1 and 4.2; the top exterior power is scalar.
For the refinement identity, use the weighted variance formula
on each complement . After multiplication by , the coefficient contributed to is . Combining this with the first sum in (3.5) gives , which proves (3.7). □
Removing the phases
We recall two standard matrix inequalities in the precise forms needed here. If is invertible and are its singular values, Weyl’s multiplicative inequality gives
Equality holds if and only if is normal. To see the equality assertion, apply strict convexity of to the majorization of logarithms of eigenvalue moduli by logarithms of singular values. Equality identifies the two multisets; (4.3) then implies normality. See [2] for multiplicative majorization.
For positive definite , the Araki–Lieb–Thirring inequality is
For , equality holds if and only if . We use Araki’s stronger log-majorization theorem [1]; a short justification of the equality condition is useful. For fixed , that theorem makes
nondecreasing on . The function is real analytic there. Equality in (4.7) says , so is constant first on and then on . In particular,
Functional calculus now gives . This argument uses only the log-majorization theorem; see also Hiai [5] for a careful treatment of equality in related norm inequalities.
Lemma 4.3. Put and . For all and ,
For , equality holds if and only if all have the same argument.
Proof. Write , with diagonal unitary, and choose a diagonal unitary with . The matrix is similar to . Its singular values are the eigenvalues of . Apply (4.6) to .
For , equality is equivalent to normality of , by Lemma 4.2. Since and , this is equivalent to , and hence to . Every off-diagonal entry of is positive. Therefore is scalar, as required. □
Proof of Theorem 3.1 and Corollary 3.2. For , apply (4.7) to and in (4.8). Since
we obtain
For and , equality also requires . Neighboring index sets and have a nonzero connecting entry in , so for every pair. Lemma 4.3 supplies the common argument. At , only the latter condition remains.
For , the diagonal of a positive matrix is majorized by its eigenvalues. Concavity therefore gives
For completeness, diagonalize unitarily and apply scalar Jensen to each diagonal entry. Strict concavity shows that equality forces to be diagonal. When , its entries joining neighboring index sets are nonzero, so the inequality is strict.
To prove that the coefficient of cannot be reduced, take for and otherwise, and let . The nonzero eigenvalues in the limit are and copies of . Thus the left-hand side tends to , while every other monomial on the right tends to zero.
For equal charges, the coefficients depend only on , giving (3.3). The preceding limiting configuration proves sharpness for . For , take all equal: the eigenvalues are and copies of , and
The case is (3.4). □
Remark 4.4 (Comparison with product majorization). Borcea’s Theorem 3 [3], applied to products of equilibrium points and the convex function , yields, after inversion using (3.4),
Indeed, normalize the charges to total mass one. The coefficient of a product indexed by in the forward inequality is ; products containing the distinguished zero vanish. Set and divide by the total product of the equilibrium points. This gives (29).
For the product identity suffices.
At , (29) and (7) coincide. For and , the coefficient in (29) exceeds ours by
Thus the endpoint is already a consequence of product majorization, while the new coefficients give a strict improvement away from that endpoint.
Normal powers and equality in higher moments
A variance identity
The next identity explains why the positive similarity improves every even-moment bound, not just the quadratic one.
Lemma 5.1. If , then
Proof. Expand using to obtain (30). The two single-rank terms have the same trace because is symmetric and is real. For (31), use in . □
Put
This matrix is symmetric, , and . Moreover,
For this is immediate. The induction step follows from :
Lemma 5.1 therefore proves (17). Its right-hand side is nonnegative by Cauchy–Schwarz, with equality precisely when all coincide. Together with Schur’s inequality, this proves the sandwich (16).
Suppose now that for every . Then and, by symmetry, , so commutes with . Consequently
commutes with both and , hence with . Since is a polynomial in (interpolate conjugation on its finite set of distinct diagonal entries), commutes with and therefore with . In particular, is normal. Conversely, normality of gives equality between the endpoints of (16), so (17) vanishes. This proves part (1) of Theorem 3.4.
Divided differences
For a polynomial , write
Lemma 5.2. For ,
Furthermore,
Proof. All generating series below may be read formally. Set
The recurrence (15) gives
Indeed, summing the recurrence first gives ; summing this at with weights identifies the parenthesis as .
The rank-one inverse formula gives
Taking the divided difference in of (35) and comparing coefficients proves (33), also when .
Finally,
Use , , and (5.4). The diagonal terms cancel, leaving (5.5). □
Completion of Theorem 3.4. For any matrix , normality of is equivalent to
One implication follows by commuting with . For the other, if is normal, its adjoint is a polynomial in , by the spectral theorem and interpolation on the finite spectrum. Since commutes with , it commutes with ; taking adjoints gives (5.7).
If the are collinear, itself is normal. Otherwise, partition the indices according to the lines through the origin containing . There are at least two classes. For indices in different classes, and . Thus (5.5) and (5.7) force . Any two indices in the same class can be joined through an index in another class, so all coincide. The converse follows from part (1) and when the variance vanishes. Finally, is monic of degree , so a level set of contains at most distinct points. □
Remark 5.3 (Polynomial functions). The same argument applies to an arbitrary polynomial . Put . Then
The matrix is normal exactly when the coincide; is normal exactly when this holds or the are collinear. To include the constant term in the norm calculation, use in Lemma 5.1. The vector differs from by a constant multiple of , which does not change the variance. The commutator formula follows by linearity, and the finite-spectrum argument above applies with in place of .
The quartic bound and higher-order examples
Proof of Corollary 3.5. With , one has . Directly,
Lemma 5.1 gives both right-hand sides of (19).
It remains to identify equality. If is constant, the monic quadratic takes that value at at most two distinct points. For two such points ,
Thus and lie on opposite rays. The alternative in part (2) of Theorem 3.4 therefore adds no noncollinear configurations when . Part (1) gives exactly (20) for equality in the unbalanced bound. □
At higher orders, noncollinear equality does occur.
Example 5.4 (A noncollinear sixth-moment equality). Take , , and
Then . For , put . Its determinant is , so Cayley–Hamilton gives
Consequently is scalar, although is not normal. Hence equality holds throughout (16) for . In polynomial terms, this is the cubic .
Example 5.5 (Regular polygons). Let and , where and . The determinant identity in Lemma 4.1 reduces to
Indeed, if , then . Thus . For this gives noncollinear equality at order and at every positive multiple of that order.
Generating functions and stability
A closed generating determinant
All the quadratic exterior refinements can be collected in one rank-two determinant.
Proposition 6.1. For , set
Then
The inequality holds coefficient by coefficient as a polynomial in . For any fixed , equality holds if and only if the are collinear.
Proof. The coefficient of in is . Theorem 3.3 proves the inequality and its equality statement (with the one-dimensional case automatic).
For the determinant formula, write , , and . The matrix is similar to , and
The rank-two determinant identity therefore reduces the quotient by to the determinant of
Here . The displayed determinant is , as claimed. □
A quantitative collinearity estimate
Equality in the reciprocal Schoenberg inequality detects a line through the origin. Its deficit also controls the squared distance to the nearest such line. Keep from (18), and define
Proposition 6.2. Let
Then
Also,
No optimality is asserted for the constants in (38).
Proof. Put , so . Straightforward rank-one products give
Expanding gives . Expanding the square of the commutator proves the second identity in (39).
In a unitary Schur basis write , where is diagonal and is strictly upper triangular. Then , and
Since , it follows that . This proves the first two estimates.
Finally, minimizing the elementary trigonometric expression defining gives . The second term in is nonnegative because , and hence
Cauchy–Schwarz gives , so (21) implies . Substitution proves the last estimate in (38). □
AI Declaration: Artificial intelligence (AI) tools were used to test the inequalities in this paper and to provide examples. An LLM also provided feedback on an earlier draft, pointed out several extensions to its original results, and assisted in re-writing of this paper. The author assumes full responsibility for the accuracy of all results presented here.
References
References
- [1]H. Araki, On an inequality of Lieb and Thirring, Lett. Math. Phys. 19 (1990), 167–170. doi:10.1007/BF01045887.DOI
- [2]R. Bhatia, Matrix Analysis, Graduate Texts in Mathematics, vol. 169, Springer, New York, 1997.
- [3]J. Borcea, Equilibrium points of logarithmic potentials induced by positive charge distributions. I. Generalized de Bruijn–Springer relations, Trans. Amer. Math. Soc. 359 (2007), 3209–3237. arXiv:math/0601519.arxiv.org/abs/math/0601519
- [4]W.-S. Cheung and T.-W. Ng, A companion matrix approach to the study of zeros and critical points of a polynomial, J. Math. Anal. Appl. 319 (2006), 690–707. doi:10.1016/j.jmaa.2006.06.071.
- [5]F. Hiai, Log-majorization and matrix norm inequalities with application to quantum information, Acta Sci. Math. (Szeged) 90 (2024), 527–549. arXiv:2402.16067.arxiv.org/abs/2402.16067
- [6]O. Kushel and M. Tyaglov, Circulants and critical points of polynomials, J. Math. Anal. Appl. 439 (2016), 634–650. doi:10.1016/j.jmaa.2016.03.005.DOI
- [7]S. M. Malamud, Inverse spectral problem for normal matrices and the Gauss–Lucas theorem, Trans. Amer. Math. Soc. 357 (2005), 4043–4064. arXiv:math/0304158.DOI
- [8]R. Pereira, Differentiators and the geometry of polynomials, J. Math. Anal. Appl. 285 (2003), 336–348. doi:10.1016/S0022-247X(03)00465-7.DOI
- [9]I. J. Schoenberg, A conjectured analogue of Rolle’s theorem for polynomials with real or complex coefficients, Amer. Math. Monthly 93 (1986), 8–13.DOI
- [10]Q. Tang and T. Zhang, Sharp Schoenberg type inequalities and the de Bruijn–Sharma problem, preprint, arXiv:2508.10341v3, 2025.arxiv.org/abs/2508.10341