theorem landscape_structure (e : NSectorEconomy N) (n : Fin N)
(hKeff : 0 < sectorEffectiveCurvature e n) :
-- The sector-n component of the landscape is strictly convex
-- on 1⊥ when K_eff_n > 0 (sub-critical regime).
-- We state the algebraic consequence: the relaxation rate is
-- strictly positive, ensuring convergence to equilibrium.
0 < sectorRelaxRate e n := by
simp only [sectorRelaxRate]
apply mul_pos
· apply mul_pos (e.hℓ n) (abs_pos.mpr _)
simp only [logCesEigenvaluePerp]
apply div_ne_zero
· simp only [neg_ne_zero]; linarith [e.hρ n]
· exact mul_ne_zero (Nat.cast_ne_zero.mpr (by have := e.hJ n; omega))
(pow_ne_zero 2 (ne_of_gt (e.hc n)))
· -- (1 - T/T*)⁺ > 0 from K_eff > 0
simp only [sectorEffectiveCurvature, effectiveCurvatureKeff] at hKeff
have hK : 0 < curvatureK (e.J n) (e.ρ n) := curvatureK_pos (e.hJ n) (e.hρ n)
have hmax : 0 < max 0 (1 - e.T n / sectorCriticalFriction e n) := by
by_contra h; push_neg at h
have := le_antisymm h (le_max_left 0 _)
rw [this, mul_zero] at hKeff; linarith
exact hmaxtheorem eigenstructure_bridge (e : NSectorEconomy N) (n : Fin N) :
-- For the block-diagonal (sector-decoupled) case, the eigenvalue of
-- the dynamics matrix M = W · ∇²V at sector n equals the sector
-- relaxation rate: μ_n = ℓ_n · |λ_⊥(n)| · (1 - T_n/T*_n)⁺.
-- This is exactly sectorRelaxRate, proving that the eigenstructure
-- bridge is definitionally built into the relaxation spectrum.
sectorRelaxRate e n =
e.ℓ n * |logCesEigenvaluePerp (e.J n) (e.ρ n) (e.c n)| *
max 0 (1 - e.T n / sectorCriticalFriction e n) := rfltheorem decay_factor_in_unit (M : ℝ) (hM : 1 < M) :
0 < 1 - 1 / M ∧ 1 - 1 / M < 1 := by
constructor
· rw [sub_pos, div_lt_one (by linarith)]; linarith
· linarith [div_pos one_pos (by linarith : (0 : ℝ) < M)]theorem steady_state_variance (M sig_new : ℝ) (hM : 1 < M) :
varianceRecurrence (M * sig_new) M sig_new = M * sig_new := by
simp only [varianceRecurrence]
have hMne : M ≠ 0 := ne_of_gt (by linarith)
field_simp
ring