Keyboard shortcuts

Press ← or → to navigate between chapters

Press ? to show this help

Press Esc to hide this help


slug: oeis-a277829-a278770-catalan-square-hankel-limits bibkey: kotesovec2016a277829a278770 doi: null url: https://oeis.org/A277829 triage: theorem motivation_gids:

  • D5/S3/Constants/Moments/CatalanSquareHankelLimits.catalan_square_hankel_log_limits

A277829 and A278770: squared-Catalan Hankel limits

Problem

Let

C_m = binomial(2*m,m)/(m+1),
D_r(n) = det[(C_(i+j+r))^2] over i,j in Fin n.

The empty determinant is one. Prove both conjuncts

lim_(n->infinity) log(D_1(n))/n^2 = 2*log(2),
lim_(n->infinity) log(D_2(n))/n^2 = 2*log(2).

These are exactly Kotesovec’s November 29, 2016 A277829 conjecture and November 28, 2016 A278770 conjecture. Their one-based programs use C_(i+j-1)^2 and C_(i+j)^2, respectively, so the zero-based shifts are exactly one and two. The source assertions were preregistered in https://github.com/the-omega-institute/trureturing/issues/9530 before the proof probes.

Motivation

The two OEIS sequences differ only by the adjacent Hankel shift, while their published conjectures assert the same exact logarithmic rate. A joint formal endpoint keeps both literal source assertions visible and lets the common moment, Gram, and asymptotic argument discharge them without identifying the two determinant sequences.

Route

The supporting module defines the scaled beta Catalan moment law and proves the literal identity integral (4*x)^m = C_m. Fubini gives the square C_m^2 as the moment of the product of two independent scaled beta variables. Weighted powers of that product coordinate are linearly independent in L2; their Gram matrices are the two shifted Hankel matrices. This proves strict positivity of every determinant for both shifts, including the size-zero determinant.

For the upper estimate, monic affine Chebyshev-T polynomials on [0,16] replace the power basis without changing the Gram determinant. Their uniform supremum bound and the positive-definite Hadamard inequality give

D_1(n) <= 4^n  * 16^(n*(n-1)/2),
D_2(n) <= 16^n * 16^(n*(n-1)/2).

For every 0<delta<8, the beta density has a positive uniform lower bound on a compact central rectangle. After transporting Chebyshev-U orthogonality to that rectangle, the proof bounds the full normalized Gram quadratic form below, not only its diagonal. A positive-semidefinite remainder and all principal minors yield one constant C>0, independent of n, such that for r=1,2

(C*delta^r)^n * ((8-delta)*pi/2)^n
  * ((8-delta)/2)^(n*(n-1)) <= D_r(n).

Taking logarithms, dividing by n^2, and moving delta to zero squeezes both rates to log 4 = 2 log 2.

Gap

The adjacent Lin note records the Catalan density and product-density route. The Simon note records the Erdos-Turan regularity criterion, monic norm asymptotics, and interval capacity. Together they give a separate ordinary classical corollary with the same two limits. The Lean proof is elementary and does not formalize that literature derivation.

The bounded prior audit described in the Kotesovec note found no explicit settlement of either exact named conjecture in its inspected scope. It is not a worldwide absence or priority certificate, and this dossier does not claim full publication or global novelty.

Falsifier

A failure of either all-n positivity statement, or a subsequence on which either displayed normalized logarithm stays away from 2*log(2), would refute the formal target. Finite numerical agreement alone cannot prove either limit.

Evidence

D5.S3.Constants.Moments.CatalanSquareHankelGrowth.catalan_square_hankel_det_positive proves both all-n positivity conjuncts, and D5.S3.Constants.Moments.CatalanSquareHankelGrowth.catalan_square_hankel_det_upper proves both explicit upper bounds. The exact named settlement is D5.S3.Constants.Moments.CatalanSquareHankelLimits.catalan_square_hankel_log_limits. Formal source hashes, statement identities, declaration types, and axiom closures are supplied by the canonical Lean report rather than duplicated in this narrative.

Triage

First tier; admission_basis: open-problem-resolution; preregistration issue 9530. The two Growth theorems have proof_shape: content and admission_basis: escape-witness. The Limits endpoint has proof_shape: content and admission_basis: open-problem-resolution. None of these declarations is a bounded enumeration, checker, numeric reduction, or certified finite instance, so utility: none is accurate.

ASSUMED-UNVERIFIED

The primary-source readings and bounded prior searches are inherited from the supplied caller and independent source-audit results. This metadata pass did not independently repeat them. The search limits, including one HTTP 403 and one HTTP 429, do not rule out unindexed literature, textbooks, private work, or later solutions. Kernel verification settles the displayed formal conjunction but does not establish worldwide priority or publication status.