The bounded-variation Lévy drift-exponent deficit vanishes on reciprocal-frequency time meshes
ProvedAvramDividend.Classical.bv_inverse_frequency_mesh_gap_tendsto_zeroasymptoticavram-classicalbounded-variationfinite-meshlevy-exponent
If the positive-time mesh width delta_n and Levy Laplace frequency theta_n satisfy delta_n*theta_n=c, with theta_n tending to infinity, then delta_n times the bounded-variation physical-drift exponent gap tends to zero. This follows by multiplying the accepted normalized exponent gap limit by c. This supplies the exact exponential rate which appears in each cell of the infinite-activity finite-mesh no-local-monotonicity proof.
Preamble
import Definitions.Def_AvramDividend_Classical_SpectrallyNegativeLevy import Theorems.Thm_AvramDividend_Classical_bv_exponent_gap_div_tendsto_zero import Mathlib.Tactic.FieldSimp import Mathlib.Tactic.Ring open AvramDividend.Classical MeasureTheory Filter open scoped NNReal ENNReal
Formal statement
theorem AvramDividend.Classical.bv_inverse_frequency_mesh_gap_tendsto_zero
{Ω : Type*} [mΩ : MeasurableSpace Ω] {P : Measure Ω}
{𝓕 : Filtration ℝ≥0 mΩ}
(X : SpectrallyNegativeLevy P 𝓕) (hBV : X.BoundedVariation)
(δ θ : ℕ → ℝ) (c : ℝ)
(hθ : Tendsto θ atTop atTop)
(hθne : ∀ n, θ n ≠ 0)
(hProduct : ∀ n, δ n * θ n = c) :
Tendsto (fun n : ℕ =>
δ n * (X.drift * θ n - X.ψ (θ n)))
atTop (nhds (0 : ℝ)) := by sorrySource
Proved bv_exponent_gap_div_tendsto_zero, Filter.Tendsto.comp and const_mul.