Covariance sum from frame correlations
ProvedConway99Formal.CubicMetric.rank_one_covarianceconway99-formal-project-20261003cubic-metricformalized-conditional-result
For finite index sets, assume and . Then . Both assumptions are explicit.
Preamble
import Mathlib namespace Conway99Formal.CubicMetric end Conway99Formal.CubicMetric set_option autoImplicit false open Conway99Formal.CubicMetric
Formal statement
theorem Conway99Formal.CubicMetric.rank_one_covariance {U T I : Type*} [Fintype U] [Fintype T]
[Fintype I] [DecidableEq I] (tau : T → I → ℝ) (h : U → T → ℝ)
(F : U → Matrix I I ℝ)
(hF : ∀ u i j, F u i j = -(∑ t, h u t * tau t i * tau t j) / 3)
(hpair : ∀ t s,
(∑ u, h u t * h u s) = 63 * (∑ k, tau t k * tau s k)) :
(∑ u, F u * F u) =
(7 : ℝ) • Matrix.of (fun i j =>
∑ t, ∑ s, (∑ k, tau t k * tau s k) ^ 2 * tau t i * tau s j) := by sorry
Source
Exact original Lean source: formalization/2026-10-03/cubic-metric/TraceForm.lean#91-202; source commit a45708acebe3f397faccb1b646be906f24f23ee5; source SHA-256 41886149af07dad84a52c54e70937fec064d5fb5f17c883bb4aea3e7935a19b3. Mechanically extracted declaration: blob/a45708acebe3f397faccb1b646be906f24f23ee5/formalization/2026-10-03/cubic-metric/TraceForm.lean#L91-L202.