Baricz–Pogány's asymptotic-coefficient positivity conjecture

What Lean checks

Original factorial products, formal division, all-degree strict positivity and the explicit lower bound are proved. The source coefficient recurrence uniquely determines the same sequence; its inequality is now an explicit endpoint. Strict compilation, an all-owned audit of 65 declarations/55 theorem declarations and all seven required endpoints, and stock-kernel replay passed. Local receipt; combined repository gate pending. Source-access and priority limits are unchanged.

This directory gives a complete proof of the unnumbered conjecture in Section 4.4 of Á. Baricz and T. K. Pogány, Probabilistic and Analytical Aspects of the Symmetric and Generalized Kaiser–Bessel Window Function, Constructive Approximation 58 (2023), 713–783, DOI 10.1007/s00365-023-09627-3.

Two independent internal mathematical reviews pass. The documented literature and public-solution-archive search found no earlier exact resolution in the material inspected. Baricz's current publication list contains an unlinked submitted follow-up on the four-parameter Kaiser–Bessel distribution; its contents were not available for review. The proof is complete, but exclusive priority is not guaranteed.

Lean status: complete. Conjectures.BariczPogany proves the full coefficient inequality, explicit lower bound, and uniqueness and positivity of the sequence defined by the source recurrence. Strict compilation, an all-owned audit of 65 declarations and seven required theorem endpoints, and independent stock-kernel replay passed. Local validation receipt. The combined repository gate remains pending.

The statement

The source defines the coefficients in the large-argument expansion

tIν(t)Iν+1(t)∼t∑n≥0αn(ν)t−n,α0(ν)=1,α1(ν)=ν+12.\frac{tI_\nu(t)}{I_{\nu+1}(t)} \sim t\sum_{n\geq0}\alpha_n(\nu)t^{-n}, \qquad \alpha_0(\nu)=1,\quad \alpha_1(\nu)=\nu+\tfrac12.

It conjectures that

αn(ν)>(ν+12)nfor every n≥2,|ν|<12.\alpha_n(\nu)>(\nu+\tfrac12)^n \qquad\text{for every }n\geq2,\quad |\nu|<\tfrac12.

The proof covers every degree and every permitted real parameter.

Why the proof works

Put a=ν+1/2a=\nu+1/2, so 0<a<10<a<1. Normalize the standard Bessel asymptotic coefficient series as a numerator NN and denominator D=1−BD=1-B. Every coefficient of NN is positive, and every positive-degree coefficient of BB is positive. Their explicit products give

D(z)=(1−az)N(z)−z2N′(z).D(z)=(1-az)N(z)-z^2N'(z).

Consequently the formal coefficient series of the desired differences is

N(z)D(z)−11−az=z2N′(z)(1−az)(1−B(z)).\frac{N(z)}{D(z)}-\frac{1}{1-az} =\frac{z^2N'(z)}{(1-az)(1-B(z))}.

Both inverse factors have nonnegative coefficients and constant term one. The numerator has a strictly positive coefficient in every degree at least two. In fact, with

cn=∏k=1n(k−a)(k+a−1)2nn!,c_n=\frac{\prod_{k=1}^n(k-a)(k+a-1)}{2^n n!},

the proof gives

αn(ν)−an≥(n−1)cn−1>0(n≥2).\alpha_n(\nu)-a^n\geq(n-1)c_{n-1}>0 \qquad(n\geq2).

These are identities of formal power series. The proof neither sums the generally divergent asymptotic expansion nor differentiates an asymptotic error term.

Files and reviews

Run the diagnostic with Python 3 using the command python results/baricz-pogany-coefficients/verify.py from the repository root. It uses exact rational arithmetic and the Python standard library. It reports 455 exact identities, 49 source fixtures, and 105 strict bounds passed. These finite checks support the written proof; they do not establish the universal statement.

The formal proof starts from the source factorial products and formal quotient. source_recurrence_unique identifies any sequence satisfying the source recurrence, and source_recurrence_conjecture proves the desired inequality directly for that sequence. No convergence claim about the asymptotic expansion is needed.