Proof of Theorem 3.1, p. 787 — pointwise Lipschitz estimates for F and G against dD
ProvedAntonelliBFSDE.BackwardForward.lipschitz_estimatesbackward-stochastic-differential-equationforward-backward-sdelipschitzp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1
Under the hypotheses of §3 and the domination of , by , let and write
- For every and every ,
- For every càdlàg version of and of and every , is integrable and, -a.s.,
These are the two estimates from which the proof of Theorem 3.1 derives the contraction property: the Lipschitz condition in , and then , .
Formalization Note The step from , to is the paper's "by absolute continuity", which rests on the domination hypothesis Dominated. In clause 1 the left side is compared in ; in clause 2 is turned into a real number, and its integrability is part of the conclusion, so the conditional expectation is not the junk value .
Preamble
import Mathlib import Definitions.Def_AntonelliBFSDE_BackwardForward_Setting import Definitions.Def_AntonelliBFSDE_BackwardForward_System open MeasureTheory open scoped NNReal ENNReal
Formal statement
namespace AntonelliBFSDE.BackwardForward
theorem lipschitz_estimates {Ω : Type*} [mΩ : MeasurableSpace Ω] (P : Measure Ω)
[IsProbabilityMeasure P] (𝓕 : Filtration ℝ≥0 mΩ) (T : ℝ≥0) (β : ℝ) (hT : 0 < T)
(husual : UsualHypotheses 𝓕 P) (IA IC : BVIntegrator 𝓕 T β) (hdom : Dominated IA IC)
(k : ℝ) (f g : ℝ≥0 → Ω → ℝ → ℝ → ℝ) (Y : Ω → ℝ) (J : ℝ≥0 → Ω → ℝ)
(hyp : StandingHyp 𝓕 P IA IC k f g Y J)
(U V U' V' : ℝ≥0 → Ω → ℝ) (hU : MemL1D P IA IC U) (hV : MemL1D P IA IC V)
(hU' : MemL1D P IA IC U') (hV' : MemL1D P IA IC V') :
(∀ ω, ∀ t ≤ T, ENNReal.ofReal |Fop IA J f U V t ω - Fop IA J f U' V' t ω| ≤
ENNReal.ofReal k * pathDist IA IC U V U' V' 0 t ω) ∧
∀ W W' : ℝ≥0 → Ω → ℝ, IsGVersion 𝓕 P IC g Y U V W → IsGVersion 𝓕 P IC g Y U' V' W' →
∀ t ≤ T, Integrable (fun ω => (pathDist IA IC U V U' V' t T ω).toReal) P ∧
∀ᵐ ω ∂P, |W t ω - W' t ω| ≤
k * (P[fun ω => (pathDist IA IC U V U' V' t T ω).toReal | 𝓕 t]) ω := by sorry
end AntonelliBFSDE.BackwardForward
Source
Antonelli, Backward-forward stochastic differential equations, Ann. Appl. Probab. 3(3) (1993), p. 787, proof of Theorem 3.1, the two displays after 'by absolute continuity, we have that'
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.