A total variation threshold for the first coordinates of uniform parking functions via Lorentzian polynomials
In the authors' words
Let be a uniform parking function of size . We determine the scale at which its first coordinates cease to be close in total variation to independent uniform variables on . The distance is comparable to , uniformly for and , and tends to one above the scale . At the critical scale , it converges to , where is the standard normal distribution function. A key ingredient is the Lorentzian property of the occupation generating polynomial of parking functions. Together with a strengthened Han inequality, this yields a relative entropy estimate for approximation by the product of the one-coordinate marginals.
Appeared: Friday, October 30. arXiv. Preprint, not yet peer-reviewed.
Authors' comment: 9 pages, 1 figure. Includes a 1.5-page section on AI use (in writing the paper and formalizing its results in Lean 4)