Degree-uniform regions for Gautschi's conjecture on subrange Jacobi polynomials

K. Castillo; L. Tertuliano da Silva; V. Botta

Degree-uniform regions for Gautschi's conjecture on subrange Jacobi polynomials

T0 review · 0 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Gautschi's conjecture proven for every degree in the positive sector

desk verdict Solid paper: resolves Gautschi's conjecture for β≥0 and for the negative wedge when c²≤3/(3+r), with careful proofs and an honest boundary; the MTP2 step is the one to scrutinize but it holds up. read the letter →

arxiv 2608.05963 v1 pith:XGU4PCLU submitted 2026-08-06 math.CA

V. Botta , K. Castillo , L. Tertuliano da Silva This is my paper · ORCID

classification math.CA MSC 33C4542C0560E1565D32

keywords Gautschi'sconjecturesubrangeJacobipolynomialszeromonotonicityorthogonalpolynomialensemblesMTP2associationPearsonidentityroot-motionidentitiesWard

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Gautschi's conjecture concerns monic orthogonal polynomials for the Jacobi weight $(1-x)^\alpha(1+x)^\beta$ restricted to a subinterval $[-c,c]$, and asserts that the endpoint ratio $[\pi_n(-c)/\pi_n(c)]^2((1-c)/(1+c))^{\beta-\alpha}$ is strictly less than $1$. Because of Gautschi's variation formula, this inequality is sufficient to conclude that every positive zero moves strictly to the right as $c$ increases. The paper proves the conjecture uniformly in the degree for $0<\alpha<\beta$, completing the entire admissible range $\beta\ge 0$, and proves it also in the negative wedge $\alpha=-r-\lambda$, $\beta=-r+\lambda$ whenever $c^2\le 3/(3+r)$. Together with earlier results this settles the conjecture for every admissible pair of parameters when $c\le\sqrt{3}/2$. The proof combines an exact boundary identity, a first-crossing lemma, and a Ward identity with positive association for an orthogonal polynomial ensemble.

What carries the argument

The argument revolves around the endpoint difference $D_n=w(c)\pi_n(c)^2-w(-c)\pi_n(-c)^2$, which the boundary identity rewrites as $\int_{-c}^{c} w'(x)\pi_n(x)^2\,dx$; the conjecture is exactly $D_n>0$. In the positive sector, the Pearson identity $d_n=\lambda+(n+\nu-1)\Sigma_n-(n+\nu)\Sigma_{n+1}$ and the root-motion formula $\partial_c\Sigma_n=a_n(U_nU_{n-1}-V_nV_{n-1})$ feed a crossing lemma: at a first zero of $d_n$ its derivative must be positive, contradicting the definition of a first crossing. In the negative wedge, a Ward identity for the ordered-chamber orthogonal polynomial ensemble, combined with the MTP$_2$ association theorem, gives the root-sum estimate $(N-ra^2)\Sigma_N < N\lambda a^2$, which supplies the root-sum bound required by the crossing lemma.

What would settle it

Compute the normalized endpoint difference $D_n/h_n$ at a fixed degree (say $n=2$ or $3$) for parameters in the remaining wedge $c^2>3/(3+r)$ using high-precision quadrature; a non-positive value would refute the conjecture in that region. Alternatively, evaluate the covariance $\mathrm{Cov}_{N,0}(S,e^{2\lambda H})$ for a small ensemble (for example $N=5$, $r=0.5$, $\lambda=0.4$, $a=0.9$); a negative value would disprove the MTP2 association hypothesis used in Lemma 5.1.

Watch

Extended reading notes

Core claim

The paper's central discovery is that Gautschi's endpoint inequality, verified pointwise at the smallest interval and prevented from crossing zero by a local transversality argument, holds for every degree uniformly in the positive sector $0<\alpha<\beta$ (Theorem 1.2). The same first-crossing mechanism, fed by a root-sum estimate derived from an MTP$_2$ orthogonal polynomial ensemble, proves the inequality throughout the negative wedge when $c^2\le 3/(3+r)$ (Theorem 1.4). Consequently, for $0<c\le\sqrt{3}/2$ the conjecture holds for every admissible pair of parameters (Corollary 1.5). The paper also establishes the case $n=1$ in full generality and shows eventual validity for each fixed admissible triple as $n\to\infty$, so at any remaining parameter point only finitely many degrees are unresolved.

Load-bearing premise

The proof in the negative wedge rests on the premise that a certain weighted sum of the roots remains non-negative on average after exponential tilting; if that positive correlation failed, the root-sum estimate and Theorem 1.4 would collapse.

Editorial extensions

If this is right

For every $n\ge 1$, $0<c<1$, and $0<\alpha<\beta$, every positive zero of $\pi_n$ moves strictly to the right as $c$ increases, by Gautschi's variation formula.
The endpoint inequality holds uniformly in the degree throughout the positive sector and in the negative-wedge slice $c^2\le 3/(3+r)$, making zero monotonicity degree-uniform in those regions.
For $0<c\le\sqrt{3}/2$, Gautschi's conjecture holds for every admissible parameter pair and every degree, closing the full admissible range.
At every fixed admissible parameter point outside the degree-uniform regions, the remaining question is reduced to finitely many degrees $n\ge 2$, since degree one is proved and eventual validity as $n\to\infty$ is established.
The Möbius reformulation converts the conjecture into the reciprocal-moment inequality $M_n>\beta/\alpha$, giving an exact finite-degree criterion in the remaining wedge.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The paper leaves implicit that a quantitative strengthening of the ensemble covariance bound used in Lemma 5.1 would close the remaining wedge and prove the conjecture for all admissible parameters.
Because eventual validity holds at every fixed parameter point, any counterexample would have to occur at a low degree; checking the finite intermediate block at each parameter point would settle the residual region.
The same MTP2-tilted Ward identity likely applies to other subrange weight families, offering a general recipe for the root-sum estimates needed in first-crossing proofs of zero monotonicity.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Candid take: this is the real thing. It resolves Gautschi's conjecture for the entire β≥0 half of the admissible range, and for the negative wedge when c²≤3/(3+r), leaving only the explicitly characterized O_c region for c>√3/2. Theorem 1.2 and Theorem 1.4 are genuine new results, not repackagings of Milovanović's criterion. The paper also proves degree-one exactly and eventual validity via weak arcsine limits.

The proof structure is sound. The boundary identity (2.2)–(2.3), the Pearson identity (3.3), the root-motion formula (3.6), and the crossing lemma (3.4) all check out. The induction in Theorem 1.2 is airtight. In the negative wedge, the Ward identity (5.7) is algebraically consistent, the boundary fluxes vanish, and the covariance nonnegativity follows from the MTP2 association theorem. I read the ensemble bound carefully and the hypotheses of [7, Theorem 4.2] are satisfied: the ordered chamber is a sublattice, the Vandermonde-square inequality is verified pairwise, the one-particle factors cancel under min/max, and S and e^{2λH} are bounded and coordinatewise increasing. The limitation in Remark 5.2 is honest and correctly states that the method just doesn't reach beyond c²=3/(3+r).

Soft spots, in proportion: (1) The MTP2 application is the most imported step; if that theorem were misapplied, Theorem 1.4 collapses. But I don't see a misapplication. A referee should still check that step first. (2) The eventual-validity result is non-effective, so the finite middle block is not bounded uniformly; the authors say this openly. (3) The degree-one proof uses a covariance inequality that is correct but a bit intricate. None of these are load-bearing.

Citation pattern is clean: earlier results by Gautschi and Milovanović are re-derived in Section 2 or cited as external criteria, not leaned on as their own. No fitted parameters; the threshold 3/(3+r) is derived. No self-citation.

Who should read this: anyone working on orthogonal polynomials, Gaussian quadrature, or zero monotonicity. It deserves a serious referee. I would send it to peer review; my expectation is acceptance after a routine check of the ensemble step.

Referee Report

0 major / 4 minor

Summary. The paper studies Gautschi's 2018 conjecture on subrange Jacobi polynomials: for -1<α<β and 0<c≤1, the endpoint ratio (1.1) is claimed to be <1. The authors introduce an equivalent endpoint difference D_n, derive exact identities, and prove the conjecture uniformly in degree for 0<α<β (Theorem 1.2). They also prove a degree-uniform family in the remaining negative wedge under c^2≤3/(3+r) (Theorem 1.4), yielding the full conjecture for all admissible parameters when c≤√3/2 (Corollary 1.5), plus degree-one and eventual-in-degree results for arbitrary parameters. The proof combines a boundary identity, Pearson and root-motion identities, a first-crossing argument, and an MTP2 ensemble Ward identity.

Significance. If correct, this is a major advance on a problem posed by Gautschi. It closes the β≥0 range completely and reduces the negative-wedge case to a parameter region Oc that is empty for c≤√3/2; for fixed parameters only finitely many degrees remain open. The paper's main tools are exact and parameter-free: no numerical evidence is used, thresholds such as 3/(3+r) are derived rather than fitted, and the positive-sector proof rests on transparent identities. The negative-wedge proof is the most delicate part; its use of the Karlin–Rinott association theorem is carefully checked, including the sublattice property, the Vandermonde-square inequality in the ordered chamber, and the Ward-identity boundary fluxes. I found no load-bearing errors.

minor comments (4)

[3.1, proof of Lemma 3.1] The sentence "Orthogonality annihilates the part containing p_n p'_n" is potentially misleading, because the integral of (1-x^2)p_n p'_n w is not zero; the nonzero contribution comes from -∫ x^2 p_n p'_n w after using ∫ p_n p'_n w = 0. A short clarification of the two integrals would improve readability.
[5, equation (5.4)] The parameter t in the density dP_{N,t} is never explicitly set equal to λ before the exponential tilting by e^{2λH} is used; stating t=λ at the definition would remove a small ambiguity.
[Figure 1, panel (b)] The panel label "c=2√5" appears to omit a slash; the caption states c=2/√5, and the label should be corrected to match.
[7, equation (7.4)] For α=0 the equivalence of (7.4) with Conjecture 1.1 degenerates to the trivial inequality 0<β; a one-line parenthetical comment would prevent confusion in that boundary case.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivations are self-contained and rest on external, re-derived, or independently proved inputs.

full rationale

The paper's principal derivations do not assume Gautschi's conjecture. Proposition 2.1 is an exact boundary identity (D_n = integral of w' pi_n^2), from which Corollaries 2.2 and 2.3 re-derive Gautschi's mixed-sign sector and Milovanović's criterion rather than importing the conjecture. Theorem 1.2 proceeds by induction with a first-crossing lemma whose hypotheses (d_{n-1}>0 and Σ_n<λ) are established at the previous degree; the base case is computed explicitly, and the Milovanović criterion only supplies an initial positivity interval. No free parameter is fitted: the threshold c^*=(β-α)/(α+β) and, in Theorem 1.4, c^2≤3/(3+r) are derived from the inequalities, the latter from N≥3 at the first possible crossing. Lemma 5.1's load-bearing association bound E_{N,λ}S≥0 comes from the external Karlin–Rinott MTP2 theorem [7] applied to an explicitly defined ensemble, and the Ward identity and determinantal representation [8] are standard; none of these inputs is equivalent to the endpoint inequality, and no cited result is by the present authors. The degree-one and weak-arcsine eventual-validity arguments are independent probabilistic/asymptotic proofs. The limitations in Remarks 5.2 and 6.2 and Section 8 explicitly mark where the ensemble estimate stops being sufficient, which is honest boundary-drawing rather than a circular rescue. Overall the derivation chain reduces to valid identities and external theorems, not to its own conclusion.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The paper introduces no fitted constants and no new physical or mathematical entities in the sense of new particles, forces, or dimensions. It relies on three published external theorems (Rakhmanov, Van Assche, Karlin-Rinott) and on the parameter hypotheses of the admissible range; the MTP2 structure of the ordered-chamber ensemble is proved in the paper and is the main imported-structure risk.

assumptions (5)

domain assumption The Jacobi weight w(x)=(1-x)^α(1+x)^β is strictly positive and bounded on [-c,c] for -1<α<β and 0<c<1, and the orthogonality measure has infinite support, so moments and Hankel matrices are positive definite.
Stated in the introduction as the admissible parameter range; used throughout for orthogonality, Cramer's rule analyticity (Section 3.1), and quadrature.
standard math Rakhmanov's ratio asymptotics: for a measure with positive continuous density on [-1,1], the ratio of successive orthonormal polynomials converges locally uniformly off the support.
Imported from [10] and used in Lemma 6.1 to identify the limiting recurrence coefficients b̂a_n→1/2, b̂b_n→0.
standard math Van Assche's weak convergence theorem for orthogonal polynomials in the class M(1,0): if recurrence coefficients tend to 1/2 and 0, p_n^2 dμ converges weakly to the arcsine law.
Imported from [11, Section 4] and used in Lemma 6.1 to obtain the weak arcsine limit (6.1).
standard math Karlin-Rinott MTP2 association theorem: for an MTP2 probability density, bounded coordinatewise increasing functions are non-negatively correlated.
Imported from [7, Theorem 4.2] and used in Lemma 5.1 to prove E_{N,λ}S ≥ 0 via the covariance with e^{2λH}.
domain assumption The ordered-chamber density (5.4) with Vandermonde squared, (1-x_i^2)^{-r} one-particle factors, and the exponential tilt is MTP2; the chamber is a sublattice and the factorization inequalities hold.
Proved in the text of Section 5; listed because the negative-wedge theorem depends on it, and it is the fragile structural input that activates the Karlin-Rinott theorem.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Degree-uniform regions for Gautschi's conjecture on subrange Jacobi polynomials." pith.science (2026). https://pith.science/paper/XGU4PCLU

@misc{pith2026260805963,
  author       = {Pith},
  title        = {Pith review of: Degree-uniform regions for Gautschi's conjecture on subrange Jacobi polynomials},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XGU4PCLU}},
  note         = {Machine review of arXiv:2608.05963}
}

abstract

Let $\pi_n$ be the monic polynomial of degree $n$ orthogonal on $[-c,c]$, $0<c\leq 1$, with respect to the Jacobi weight $(1-x)^\alpha(1+x)^\beta$, where $-1<\alpha<\beta$. Gautschi conjectured that $$ \left[ \frac{\pi_n(-c)}{\pi_n(c)} \right]^2 \left(\frac{1-c}{1+c}\right)^{\beta-\alpha} <1. $$ By his variation formula, this inequality is sufficient for every positive zero of $\pi_n$ to move to the right as $c$ increases. For $0<c<1$, we prove the conjecture, uniformly in the degree, throughout $0<\alpha<\beta$. Combined with the region $\alpha\leq 0\leq\beta$, recorded by Gautschi on the basis of an unpublished communication from Milovanovi'c, and with Milovanovi'c's published criterion, this settles the full admissible range $\beta\geq 0$. In the remaining negative wedge, writing $\alpha=-r-\lambda$ and $\beta=-r+\lambda$, we prove the conjecture whenever $$ c^2\leq\frac{3}{3+r}. $$ Consequently, it holds for every admissible pair of parameters when $0<c\leq\sqrt{3}/2$. For arbitrary admissible parameters, we also establish the degree-one case and eventual validity as $n\to\infty$. The proof combines an exact boundary identity, a first-crossing argument based on Pearson and root-motion identities, and, in the negative wedge, a Ward identity with positive association for an MTP$_2$ orthogonal polynomial ensemble. The case $c=1$ is immediate.

Figures

Figures reproduced from arXiv: 2608.05963 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 9 canonical work pages

[1]

Da Fies and M

G. Da Fies and M. Vianello,Trigonometric Gaussian quadrature on subintervals of the period, Electron. Trans. Numer. Anal.39(2012), 102–112

work page 2012
[2]

Gautschi,Orthogonal Polynomials: Computation and Approximation, Oxford University Press, Oxford, 2004

W. Gautschi,Orthogonal Polynomials: Computation and Approximation, Oxford University Press, Oxford, 2004. doi:10.1093/oso/9780198506720.001.0001

arXiv 2004
[3]

Gautschi,Sub-range Jacobi polynomials, Numer

W. Gautschi,Sub-range Jacobi polynomials, Numer. Algorithms61(2012), no. 4, 649–657. doi:10.1007/s11075-012-9556-z

work page doi:10.1007/s11075-012-9556-z 2012
[4]

Gautschi,Erratum to: Sub-range Jacobi polynomials, Numer

W. Gautschi,Erratum to: Sub-range Jacobi polynomials, Numer. Algorithms74(2017), no. 2,

work page 2017
[5]

Gautschi,Correction to: Sub-range Jacobi polynomials, Numer

W. Gautschi,Correction to: Sub-range Jacobi polynomials, Numer. Algorithms81(2019), no. 2, 771. doi:10.1007/s11075-019-00715-9

work page doi:10.1007/s11075-019-00715-9 2019
[6]

Gautschi,On the zeros of subrange Jacobi polynomials, Numer

W. Gautschi,On the zeros of subrange Jacobi polynomials, Numer. Algorithms79(2018), no. 3, 759–768. doi:10.1007/s11075-017-0459-x. DEGREE-UNIFORM REGIONS FOR GAUTSCHI’S CONJECTURE 21

work page doi:10.1007/s11075-017-0459-x 2018
[7]

Karlin and Y

S. Karlin and Y. Rinott,Classes of orderings of measures and related correlation inequalities. I. Multivariate totally positive distributions, J. Multivariate Anal.10(1980), no. 4, 467–498. doi:10.1016/0047-259X(80)90065-2

doi:10.1016/0047-259x(80)90065-2 1980
[8]

König,Orthogonal polynomial ensembles in probability theory, Probab

W. König,Orthogonal polynomial ensembles in probability theory, Probab. Surv.2(2005), 385–447. doi:10.1214/154957805100000177

doi:10.1214/154957805100000177 2005

Show all 12 references

[9]

G. V. Milovanović,On Gautschi’s conjecture on subrange Jacobi polynomials, Innovative Journal of Mathematics1(2022), no. 1, 1–6. doi:10.55059/ijm.2022.1.1/15

2022 doi
[10]

E. A. Rakhmanov,On the asymptotics of the ratio of orthogonal polynomials, Math. USSR-Sb. 32(1977), no. 2, 199–213. doi:10.1070/SM1977V032N02ABEH002377

1977 doi
[11]

Van Assche,Weak convergence of orthogonal polynomials, Indag

W. Van Assche,Weak convergence of orthogonal polynomials, Indag. Math. (N.S.)6(1995), no. 1, 7–23. doi:10.1016/0019-3577(95)98197-J. Department of Mathematics and Computer Science, São Paulo State University (UNESP), Presidente Prudente, SP, Brazil Email address:vanessa.botta@...

1995 doi
[637]

doi:10.1007/s11075-016-0257-x

doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Da Fies and M

G. Da Fies and M. Vianello,Trigonometric Gaussian quadrature on subintervals of the period, Electron. Trans. Numer. Anal.39(2012), 102–112

work page 2012

[2] [2]

Gautschi,Orthogonal Polynomials: Computation and Approximation, Oxford University Press, Oxford, 2004

W. Gautschi,Orthogonal Polynomials: Computation and Approximation, Oxford University Press, Oxford, 2004. doi:10.1093/oso/9780198506720.001.0001

arXiv 2004

[3] [3]

Gautschi,Sub-range Jacobi polynomials, Numer

W. Gautschi,Sub-range Jacobi polynomials, Numer. Algorithms61(2012), no. 4, 649–657. doi:10.1007/s11075-012-9556-z

work page doi:10.1007/s11075-012-9556-z 2012

[4] [4]

Gautschi,Erratum to: Sub-range Jacobi polynomials, Numer

W. Gautschi,Erratum to: Sub-range Jacobi polynomials, Numer. Algorithms74(2017), no. 2,

work page 2017

[5] [5]

Gautschi,Correction to: Sub-range Jacobi polynomials, Numer

W. Gautschi,Correction to: Sub-range Jacobi polynomials, Numer. Algorithms81(2019), no. 2, 771. doi:10.1007/s11075-019-00715-9

work page doi:10.1007/s11075-019-00715-9 2019

[6] [6]

Gautschi,On the zeros of subrange Jacobi polynomials, Numer

W. Gautschi,On the zeros of subrange Jacobi polynomials, Numer. Algorithms79(2018), no. 3, 759–768. doi:10.1007/s11075-017-0459-x. DEGREE-UNIFORM REGIONS FOR GAUTSCHI’S CONJECTURE 21

work page doi:10.1007/s11075-017-0459-x 2018

[7] [7]

Karlin and Y

S. Karlin and Y. Rinott,Classes of orderings of measures and related correlation inequalities. I. Multivariate totally positive distributions, J. Multivariate Anal.10(1980), no. 4, 467–498. doi:10.1016/0047-259X(80)90065-2

doi:10.1016/0047-259x(80)90065-2 1980

[8] [8]

König,Orthogonal polynomial ensembles in probability theory, Probab

W. König,Orthogonal polynomial ensembles in probability theory, Probab. Surv.2(2005), 385–447. doi:10.1214/154957805100000177

doi:10.1214/154957805100000177 2005