Proposition 2.3 — is the value of an optimal stopping problem
DisprovedReflectedBSDE.Existence.proposition_2_3Let satisfy (i)–(iv) and , with moreover . Let be a solution of the reflected BSDE satisfying (v)–(viii) and (v′). Then for each ,
where is the set of stopping times with , and the essential supremum is taken relative to .
The solution of the reflected BSDE is the value process of an optimal stopping problem: stop at and receive the obstacle, or wait until and receive , with running reward in between.
Formalization Note Two hypotheses are added. The first is : under (iv) alone only is square integrable, so need not be integrable, and Lean's conditional expectation of a non-integrable variable is . The paper's Remark 3.2 shows that this is no loss of generality. The second is (v′), which follows from (v) by Corollary 3.3 and which the paper's proof uses ("From (v) and (v′)"). The essential supremum is the predicate IsEssSupOver relative to .
import Mathlib import Definitions.Def_Peng1990_SMP_Stochastic import Definitions.Def_ReflectedBSDE_Existence_Skorohod import Definitions.Def_ReflectedBSDE_Existence_Setting import Definitions.Def_ReflectedBSDE_Existence_Solution import Definitions.Def_ReflectedBSDE_Existence_OptimalStopping open MeasureTheory ProbabilityTheory Set open scoped NNReal ENNReal open Peng1990.SMP
namespace ReflectedBSDE.Existence
/-- Proposition 2.3 (p. 705): for a solution of (v)–(viii) (hence of (v′)), and with the obstacle
in `𝒮²` (Remark 3.2), for each `t ∈ [0, T]`,
`Y_t = ess sup_{v ∈ 𝒯_t} E[∫ₜᵛ f(s, Y_s, Z_s) ds + S_v 1_{v<T} + ξ 1_{v=T} | 𝓕_t]`. -/
theorem proposition_2_3 {d : ℕ} {Ω : Type*} [MeasurableSpace Ω] {P : Measure Ω}
[IsProbabilityMeasure P] {B : ℝ≥0 → Ω → Fin d → ℝ} (hB : IsStdBrownian P B)
(T : ℝ≥0) (L : ℝ) (ξ : Ω → ℝ) (f : ℝ≥0 → Ω → ℝ → (Fin d → ℝ) → ℝ) (S : ℝ≥0 → Ω → ℝ)
(hdata : IsStandardData (augmentedFiltration P hB) P T L ξ f S)
(hS2 : IsS2 (augmentedFiltration P hB) P T S)
(Y : ℝ≥0 → Ω → ℝ) (Z : ℝ≥0 → Ω → Fin d → ℝ) (K : ℝ≥0 → Ω → ℝ)
(hZ : SatisfiesV (augmentedFiltration P hB) P T Z)
(hsol : SolvesRBSDE (augmentedFiltration P hB) P T B ξ f S Y Z K)
(hYK : SatisfiesVPrime (augmentedFiltration P hB) P T Y K) :
∀ t ≤ T, IsEssSupOver P (augmentedFiltration P hB t)
(fun v : stoppingTimesBetween (augmentedFiltration P hB) t T =>
condExp (augmentedFiltration P hB t) P (stoppingReward T ξ f S Y Z t v))
(Y t) := by sorry
end ReflectedBSDE.Existence
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.