Theorem B.1 — Study–Cayley determinant comparison
OpenGinibrePermanent977.Anticoncentration.Supplement.theorem_B_1anticoncentrationp2o-batch-stoc27sp2o-gran-per-chapterp2o-plan-paperp2o-v1probabilityrandom-matrix-theory
Let be a standard quaternionic Ginibre matrix, whose real coordinates have variance . Let be the Study determinant and
with factors in increasing row order. For every ,
Equivalently, the Laplace expectation of the squared Study determinant is at least that of the squared Cayley magnitude for every nonnegative test parameter. This compares two different quaternionic determinants in distributional order, without asserting pointwise domination or equality of their laws.
Differences from the source:
- encoding: Laplace comparison for s ≥ 0 → published LaplaceOrder for s > 0; reason: at s = 0 both bounded integrands equal 1 under probability laws
- encoding: the signed row-ordered quaternionic Cayley determinant → the published generic cdetK specialized to Scalar.quaternion; reason: its Leibniz sum needs only a ring and is precisely the Appendix B definition, despite the older base prose describing only its commutative uses
Preamble
import Mathlib import Definitions.Def_GinibrePermanent977_Anticoncentration_Model open MeasureTheory ProbabilityTheory open scoped Quaternion
Formal statement
namespace GinibrePermanent977.Anticoncentration.Supplement
/-- Theorem B.1, p.17. Sdet(G)^2 ≤Lt |Cdet(G)|^2; Cayley factors are row-ordered. -/
theorem theorem_B_1 (n : ℕ) (hn : 1 ≤ n) :
StochasticOrders.LaplaceTransform.LaplaceOrder
(ginibre .quaternion n n) (ginibre .quaternion n n)
(fun G => studyDet G ^ 2) (fun G => ‖cdetK G‖ ^ 2) := by sorry
end GinibrePermanent977.Anticoncentration.Supplement
Source
Frederic Koehler, Pui Kuen Leung, STOC 2027 submission #977 (October 2026), Theorem B.1, p.17 (PDF SHA256 e25737f1d5e3492905f8eeaa41d92c55d95f73622a3b5846cc87aac6a346d4ec)