Theorem 7.2 — the inner essential infimum is real valued
OpenReflectedBSDE.StoppingControl.inner_essInf_existsLet be a standard -dimensional Brownian motion on a probability space with augmented natural filtration . Let the data satisfy the standing assumptions (i)–(iv) of §2 with a.s., let , let be Lipschitz in (iii), and let be concave on for each , . Let be the conjugate of , and let be the class of bounded progressively measurable processes with . For each let be continuous Itô integrals , which define
the solution of , , and let
be the payoff of Theorem 7.2, for and (stopping times with ).
Then for every and every the family
has a real-valued essential infimum relative to , i.e. an -measurable real random variable that is a.s. below every member of the family and a.s. above every -measurable real lower bound of the family.
This is the existence of the inner essential infimum appearing in the last expression of Theorem 7.2, , as a real random variable.
Formalization Note Essential infima are relative to and real valued (the published predicate MultiperiodRisk.Bellman.IsEssInf); this is item 4 of the formal goal theorem minimax_representation.
import Mathlib import Definitions.Def_Peng1990_SMP_Stochastic import Definitions.Def_MultiperiodRisk_Bellman_EssInf import Definitions.Def_ReflectedBSDE_StoppingControl_Setting import Definitions.Def_ReflectedBSDE_StoppingControl_Control
namespace ReflectedBSDE.StoppingControl
open MeasureTheory ProbabilityTheory Filter Topology
open scoped ENNReal NNReal
/-- Theorem 7.2 (El Karoui et al. 1997, p. 725), the inner essential infimum of the
sup–inf expression is a real random variable. Let the data `(ξ, f, S)` satisfy (i)–(iv) with
`S ∈ 𝒮²`, `f` Lipschitz and concave in `(y, z)`, and for each `(β, γ) ∈ 𝒜` let `J^{β,γ}` be
continuous Itô integrals of `γ`. Then for every `t ∈ [0, T]` and every `v ∈ 𝒯_t`, the family
`{E[Φ(t, v, β, γ) | 𝓕_t] : (β, γ) ∈ 𝒜}` has a real-valued `𝓕_t`-essential infimum. -/
theorem inner_essInf_exists {Ω : Type*} [MeasurableSpace Ω] {d : ℕ} {P : Measure Ω}
[IsProbabilityMeasure P] {B : ℝ≥0 → Ω → Fin d → ℝ} (hB : Peng1990.SMP.IsStdBrownian P B)
(T : ℝ≥0) (ξ : Ω → ℝ) (f : ℝ≥0 → Ω → ℝ → (Fin d → ℝ) → ℝ) (S : ℝ≥0 → Ω → ℝ)
(hdata : StandingData (augFiltration hB) P T ξ f S)
(hLip : ∃ K : ℝ, IsLipschitzCoeff P T f K)
(hS2 : InS2 (augFiltration hB) P T S)
(hconc : ∀ ω, ∀ t ≤ T, ConcaveOn ℝ Set.univ (fun p : ℝ × (Fin d → ℝ) => f t ω p.1 p.2))
(J : Control Ω d → Fin d → ℝ≥0 → Ω → ℝ)
(hJ : ∀ c ∈ admissible (augFiltration hB) P T f,
IsContItoIntegral (augFiltration hB) P T B c.2 (J c)) :
∀ t ≤ T, ∀ v ∈ stoppingTimesFrom (augFiltration hB) T t, ∃ X : Ω → ℝ,
MultiperiodRisk.Bellman.IsEssInf P (augFiltration hB t)
{X' | ∃ c ∈ admissible (augFiltration hB) P T f,
X' = P[payoff T ξ f S c (J c) t v | augFiltration hB t]} X := by sorry
end ReflectedBSDE.StoppingControl