Cross-terminal projective consistency of full bounded stopped-future-increment pushforward laws
ProvedAvramDividend.Classical.esscher_bounded_stop_future_increment_terminal_consistency_mapavram-classicalbounded-stoppingesscherfinite-terminalmeasure-mapprojective-consistencystopped-future-increment
For a genuine bounded Lévy stopping time τ≤H, a stopped-history event A, a lag u≥0 and any deterministic Esscher terminal times T₁,T₂≥H+1+u, the full pushforward law of the stopped future increment X_(τ+u)−Xτ under Q_T₁ restricted to A equals its law under Q_T₂ restricted to A. This is finite-horizon projective consistency for every measurable future-increment test, with no global Esscher path law or required ordering between T₁ and T₂.
Preamble
import Mathlib.MeasureTheory.Integral.Lebesgue.Map import Theorems.Thm_AvramDividend_Classical_esscher_bounded_stop_future_increment_stopped_likelihood_map_law open AvramDividend.Classical MeasureTheory Filter Topology open scoped NNReal ENNReal
Formal statement
theorem AvramDividend.Classical.esscher_bounded_stop_future_increment_terminal_consistency_map
{Ω : Type*} [mΩ : MeasurableSpace Ω] {P : Measure Ω}
[IsProbabilityMeasure P] {𝓕 : Filtration ℝ≥0 mΩ}
(X : SpectrallyNegativeLevy P 𝓕) (θ : ℝ) (hθ : 0 ≤ θ)
(τ : Ω → ℝ≥0)
(hstop : IsStoppingTime 𝓕 (fun ω => (τ ω : WithTop ℝ≥0)))
(H : ℝ≥0) (hbound : ∀ ω, τ ω ≤ H)
(u T₁ T₂ : ℝ≥0) (hT₁ : H + 1 + u ≤ T₁)
(hT₂ : H + 1 + u ≤ T₂)
(A : Set Ω)
(hA : ∀ s : ℝ≥0,
MeasurableSet[𝓕 s] (A ∩ {ω : Ω | τ ω ≤ s})) :
let QT₁ : Measure Ω := P.withDensity (fun ω => ENNReal.ofReal
(Real.exp (θ * X.X T₁ ω - (T₁ : ℝ) * X.ψ θ)))
let QT₂ : Measure Ω := P.withDensity (fun ω => ENNReal.ofReal
(Real.exp (θ * X.X T₂ ω - (T₂ : ℝ) * X.ψ θ)))
Measure.map (fun ω : Ω => X.X (τ ω + u) ω - X.X (τ ω) ω)
(QT₁.restrict A) =
Measure.map (fun ω : Ω => X.X (τ ω + u) ω - X.X (τ ω) ω)
(QT₂.restrict A) := by sorrySource
Apply canonically accepted mixed bounded-stop future-increment map law dd39f9a0 separately at T₁ and T₂. Each side equals the same realised stopped-history mass QStop(A) times the deterministic Q_u increment law; no independence under QStop or Wiener–Hopf theorem.