Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Example 1 — representations of U and V

Proved
AntonelliBFSDE.SingularExample.example_1_formulas

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

backward-stochastic-differential-equationconditional-expectationp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Suppose (U,V)(U,V)(U,V) solves Antonelli's system (3.6)–(3.7) on a positive horizon TTT, with J0>0J_0>0J0​>0 and Y>0Y>0Y>0 almost surely and the filtered-space and integrability assumptions of Example 1. Then the two processes can be expressed in terms of each other:

Ut=et(J0+∫0te−s∣Vs∣ ds),Vt=EP ⁣[eT−tY+∫tTes−tUs ds∣Ft].U_t=e^t\left(J_0+\int_0^t e^{-s}|V_s|\,ds\right),\qquad V_t=E_P\!\left[e^{T-t}Y+\int_t^T e^{s-t}U_s\,ds\mid\mathcal F_t\right].Ut​=et(J0​+∫0t​e−s∣Vs​∣ds),Vt​=EP​[eT−tY+∫tT​es−tUs​ds∣Ft​].

The identities hold dt⊗Pdt\otimes Pdt⊗P-almost everywhere. The random variable inside the second conditional expectation is integrable for each t≤Tt\le Tt≤T. These formulas isolate the positivity mechanism used in Example 1.

Formalization Note The source writes pointwise process formulas; the solution itself is an L1(dt⊗P)L^1(dt\otimes P)L1(dt⊗P) equivalence class, so the formal statements assert zero L1L^1L1 residual and a jointly measurable conditional-expectation version. The paper's “positive” is read as strictly positive almost surely; its §3 integrability condition on the constant input process Jt=J0J_t=J_0Jt​=J0​ is stated as J0∈L1(P)J_0\in L^1(P)J0​∈L1(P).

Preamble
import Mathlib
import Definitions.Def_AntonelliBFSDE_SingularExample_Setting

open MeasureTheory
open scoped NNReal ENNReal
Formal statement
namespace AntonelliBFSDE.SingularExample

/-- The two displayed representations in Example 1, p. 790, read as
`dt ⊗ P`-almost-everywhere identities. -/
theorem example_1_formulas {Ω : Type*} [mΩ : MeasurableSpace Ω]
    (P : Measure Ω) [IsProbabilityMeasure P]
    (𝓕 : Filtration ℝ≥0 mΩ) (T : ℝ≥0) (hT : 0 < T)
    (J₀ Y : Ω → ℝ) (U V : ℝ≥0 → Ω → ℝ)
    (husual : AntonelliBFSDE.BackwardForward.UsualHypotheses 𝓕 P)
    (hJmeas : Measurable[𝓕 0] J₀) (hJint : Integrable J₀ P)
    (hJpos : ∀ᵐ ω ∂P, 0 < J₀ ω)
    (hYmeas : Measurable[𝓕 T] Y) (hYint : Integrable Y P)
    (hYpos : ∀ᵐ ω ∂P, 0 < Y ω)
    (hsol : IsDtSystemSolution 𝓕 P T J₀ (fun u v => u + |v|)
      (fun u v => u + v) Y U V) :
    L1NormDt P T (fun t ω => U t ω -
      Real.exp (t : ℝ) * (J₀ ω +
        ∫ s in Set.Ioc (0 : ℝ) (t : ℝ),
          Real.exp (-s) * |V s.toNNReal ω|)) = 0 ∧
    (∀ t ≤ T, Integrable
      (fun ω => Real.exp ((T : ℝ) - (t : ℝ)) * Y ω +
        ∫ s in Set.Ioc (t : ℝ) (T : ℝ),
          Real.exp (s - (t : ℝ)) * U s.toNNReal ω) P) ∧
    ∃ W : ℝ≥0 → Ω → ℝ,
      AntonelliBFSDE.Backward.IsCondExpVersion 𝓕 P T
        (fun t ω => Real.exp ((T : ℝ) - (t : ℝ)) * Y ω +
          ∫ s in Set.Ioc (t : ℝ) (T : ℝ),
            Real.exp (s - (t : ℝ)) * U s.toNNReal ω) W ∧
      L1NormDt P T (fun t ω => W t ω - V t ω) = 0 := by sorry

end AntonelliBFSDE.SingularExample
Source
Antonelli, Backward-forward stochastic differential equations, Ann. Appl. Probab. 3(3) (1993), p. 790, Example 1, two displays after (3.7), https://doi.org/10.1214/aoap/1177005363
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me